Demographic information
Note: We set the following quotas:
- 50% women (50% men).
- 60% White (15% Black, 15% Asian, 7% Mixed, 3% Other).
I report summary statistics for:
- Age.
- Role.
- Self-Power.
- Company’s business.
Summary stats for likert dependent variables
| Item in Qualtrics | Item Text | Merrick’s Shorthand |
|---|---|---|
| Q8 | To what extent do you agree or disagree with the following statement: “Being responsible in using gen AI is important to gain business benefits | ResponsibleBusiness |
| Q13_1 | On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly? - Business pressures (e.g., speed to market) | BusinessPressure |
| Q13_2 | On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly? - Concerns about being labeled a “troublemaker” | Concerns |
| Q13_3 | On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly? - Discomfort raising issues around responsibility | Discomfort |
| Q13_4 | On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly? - Other | Other |
| Q15_1 | To what extent do you agree with the following‚ - I have suggestions for my supervisor about how to use gen AI more responsibly | Suggestions for use |
| Q15_2 | To what extent do you agree with the following, - I feel comfortable volunteering suggestions to my supervisor about how to use gen AI more responsibly | Feel comfortable volunteering suggestions |
| Q15_3 | To what extent do you agree with the following‚ - My supervisor asks and/or encourages me to think about and take actions to help the team be more responsible in using gen AI. | Take actions |
Summary statistics across dependent variables
Summary stats
Correlations
Age
Distribution
##
## 18-24 YO 25-34 YO 35-44 YO 45-54 YO 55-64 YO 65+ YO
## 25 118 90 47 17 5
Main effects
Post hoc
Q15_2 = Feel comfortable volunteering suggestions.
Role
Distribution
##
## Product manager/Lead Engineer/\nData scientist Product marketing/\nbusiness development Other Product designer/\nUX/content designer
## 149 23 72 33 25
Main effects
Ladder
Distribution
##
## 1. 1st Rung 2. 2nd Rung 3. 3rd Rung 4. 4th Rung 5. 5th Rung 6. 6th Rung 7. 7th Rung 8. 8th Rung 9. 9th Rung 10. 10th Rung
## 30 13 43 43 36 47 40 31 11 8
Main effects
Company size
Distribution
##
## 1. 1-49 2. 50-499 3. 499-1000 4. 1000-4999 5. 5000+ 99. Don't know
## 91 92 40 28 46 2
Main effects
Post hoc
Q15_3 = “Take actions”.
Industry
Distribution
##
## IT Manufacturing Healthcare Transportation Finance Retail Education Other Telecommunications Arts Agriculture Hospitality Automotive Construction CRM
## 60 48 17 14 31 37 13 26 5 20 6 9 2 8 6
Main Effects
- Got error “not enough observations”, which occurred because some industries had “0”.
Gender and Race Main Effects
Gender
Race
Formal training
genaiclean <- genaiclean %>%
mutate(formaltrainingtext = case_when(
Q10 == 1 ~ "Yes",
Q10 == 2 ~ "No"))
means_sd_table(
condition_text = "formaltrainingtext",
condition_num = "Q10",
exp_variables = dvs,
exp_variables_labels = dvlabels,
data = genaiclean, means_comp = "t-test"
)## `summarise()` has grouped output by 'variable'. You can override using the `.groups` argument.Foundational Models
External Proprietary Models
External Open Source Gen AI models
Percentages
Q1: For what types of work tasks / purpose(s) do you use gen AI? Check all that apply.
Q3: What gen AI tools/models do you use? Check all that apply.
Q8: “Being responsible in using gen AI is important to gain business benefits”
Q9: Do you take any of the following actions when using gen AI tools? Check all that apply.
Q10: “Do you have any formal training in using gen AI responsibly?”
##
## Yes No
## 0.3201 0.6766Q11: Does your organization have… (Check all that apply)
Q12: What types of challenges have you or your team faced in regards to using gen AI responsibly?
Q13: “On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly?”
Q15: To what extent do you agree with the following?
Output sheet is available here.
Results below are more clearly represented (and color-coded) in the hyperlinked spreadsheet.
https://docs.google.com/spreadsheets/d/109SUQaJhzkFDCEKAwN5ybne99zEu0WM7D0K00Ah9VcM/edit?usp=sharing
DVs: Q9: Do you take any of the following actions when using Gen AI tools? Check all that apply
Q11 Pred: Leadership that has expressed commitment to responsible AI
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.885 0.165 -5.37 0.000000079 ***
## orghave_leadershipYes 1.358 0.247 5.50 0.000000039 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 381.54 on 301 degrees of freedom
## AIC: 385.5
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.721 0.209 -8.24 < 0.0000000000000002 ***
## orghave_leadershipYes 1.316 0.277 4.74 0.0000021 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 319.77 on 301 degrees of freedom
## AIC: 323.8
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.238 0.180 -6.90 0.0000000000053 ***
## orghave_leadershipYes 0.866 0.256 3.39 0.0007 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 358.71 on 301 degrees of freedom
## AIC: 362.7
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.387 0.153 -2.53 0.0113 *
## orghave_leadershipYes 0.628 0.236 2.66 0.0078 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 411.71 on 301 degrees of freedom
## AIC: 415.7
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.885 0.165 -5.37 0.000000079 ***
## orghave_leadershipYes 0.933 0.243 3.84 0.00013 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 388.26 on 301 degrees of freedom
## AIC: 392.3
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.185 0.249 -8.79 <0.0000000000000002 ***
## orghave_leadershipYes 0.799 0.334 2.39 0.017 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 241.71 on 301 degrees of freedom
## AIC: 245.7
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.206 0.178 -6.78 0.000000000012 ***
## orghave_leadershipYes 0.834 0.255 3.28 0.0011 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 361.16 on 301 degrees of freedom
## AIC: 365.2
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.372 0.187 -7.35 0.00000000000019 ***
## orghave_leadershipYes 1.196 0.259 4.62 0.00000388250274 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 351.56 on 301 degrees of freedom
## AIC: 355.6
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.996 0.169 -5.90 0.0000000037 ***
## orghave_leadershipYes -1.246 0.347 -3.59 0.00033 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 286.57 on 301 degrees of freedom
## AIC: 290.6
##
## Number of Fisher Scoring iterations: 4Q11 Pred: Responsible or ethical AI principles
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.145 0.177 -6.46 0.0000000001021 ***
## orghave_responsibleYes 1.873 0.258 7.25 0.0000000000004 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 355.13 on 301 degrees of freedom
## AIC: 359.1
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.569 0.201 -7.82 0.0000000000000055 ***
## orghave_responsibleYes 1.012 0.272 3.73 0.00019 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 329.19 on 301 degrees of freedom
## AIC: 333.2
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.452 0.193 -7.51 0.000000000000059 ***
## orghave_responsibleYes 1.250 0.262 4.77 0.000001857187748 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 346.55 on 301 degrees of freedom
## AIC: 350.6
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.348 0.154 -2.26 0.024 *
## orghave_responsibleYes 0.519 0.234 2.22 0.027 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 413.91 on 301 degrees of freedom
## AIC: 417.9
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.880 0.167 -5.29 0.00000013 ***
## orghave_responsibleYes 0.896 0.242 3.70 0.00022 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 389.33 on 301 degrees of freedom
## AIC: 393.3
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.436 0.279 -8.74 < 0.0000000000000002 ***
## orghave_responsibleYes 1.198 0.350 3.43 0.00061 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.5 on 302 degrees of freedom
## Residual deviance: 234.9 on 301 degrees of freedom
## AIC: 238.9
##
## Number of Fisher Scoring iterations: 5Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.275 0.183 -6.95 0.0000000000037 ***
## orghave_responsibleYes 0.947 0.256 3.70 0.00022 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 358.05 on 301 degrees of freedom
## AIC: 362.1
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.415 0.191 -7.40 0.00000000000013 ***
## orghave_responsibleYes 1.244 0.260 4.78 0.00000176021321 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 349.79 on 301 degrees of freedom
## AIC: 353.8
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.853 0.166 -5.15 0.00000026 ***
## orghave_responsibleYes -1.864 0.401 -4.65 0.00000334 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 272.22 on 301 degrees of freedom
## AIC: 276.2
##
## Number of Fisher Scoring iterations: 5Q11 Pred: A policy/policies that inform the use of gen AI
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.995 0.164 -6.07 0.0000000012445 ***
## orghave_policyYes 1.769 0.260 6.81 0.0000000000098 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 362.61 on 301 degrees of freedom
## AIC: 366.6
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.447 0.185 -7.81 0.0000000000000057 ***
## orghave_policyYes 0.870 0.269 3.23 0.0012 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 332.99 on 301 degrees of freedom
## AIC: 337
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.222 0.174 -7.05 0.0000000000019 ***
## orghave_policyYes 0.904 0.257 3.52 0.00044 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 357.89 on 301 degrees of freedom
## AIC: 361.9
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.486 0.150 -3.24 0.0012 **
## orghave_policyYes 0.950 0.244 3.90 0.000098 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 403.25 on 301 degrees of freedom
## AIC: 407.2
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.815 0.158 -5.17 0.00000024 ***
## orghave_policyYes 0.850 0.245 3.47 0.00052 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 391.07 on 301 degrees of freedom
## AIC: 395.1
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.885 0.325 -8.88 < 0.0000000000000002 ***
## orghave_policyYes 1.987 0.385 5.16 0.00000025 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 215.43 on 301 degrees of freedom
## AIC: 219.4
##
## Number of Fisher Scoring iterations: 5Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.163 0.171 -6.81 0.0000000000097 ***
## orghave_policyYes 0.809 0.256 3.16 0.0016 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 361.99 on 301 degrees of freedom
## AIC: 366
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.284 0.176 -7.27 0.00000000000035 ***
## orghave_policyYes 1.108 0.258 4.30 0.00001743394308 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 354.85 on 301 degrees of freedom
## AIC: 358.9
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.891 0.160 -5.56 0.000000027 ***
## orghave_policyYes -2.191 0.485 -4.52 0.000006102 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 269.00 on 301 degrees of freedom
## AIC: 273
##
## Number of Fisher Scoring iterations: 5Q11 Pred: Clear incentives
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.485 0.132 -3.68 0.00024 ***
## orghave_incentivesYes 0.933 0.298 3.13 0.00173 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 403.24 on 301 degrees of freedom
## AIC: 407.2
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.407 0.161 -8.74 < 0.0000000000000002 ***
## orghave_incentivesYes 1.373 0.306 4.48 0.0000073 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 323.74 on 301 degrees of freedom
## AIC: 327.7
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.971 0.143 -6.77 0.000000000013 ***
## orghave_incentivesYes 0.594 0.301 1.97 0.049 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 366.56 on 301 degrees of freedom
## AIC: 370.6
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.214 0.129 -1.66 0.097 .
## orghave_incentivesYes 0.452 0.292 1.55 0.122
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 416.44 on 301 degrees of freedom
## AIC: 420.4
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.681 0.136 -5.02 0.00000051 ***
## orghave_incentivesYes 0.988 0.296 3.34 0.00085 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 391.95 on 301 degrees of freedom
## AIC: 395.9
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.003 0.198 -10.13 <0.0000000000000002 ***
## orghave_incentivesYes 0.836 0.364 2.29 0.022 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.5 on 302 degrees of freedom
## Residual deviance: 242.6 on 301 degrees of freedom
## AIC: 246.6
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.951 0.143 -6.66 0.000000000027 ***
## orghave_incentivesYes 0.574 0.301 1.91 0.057 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 368.48 on 301 degrees of freedom
## AIC: 372.5
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.911 0.142 -6.43 0.00000000012 ***
## orghave_incentivesYes 0.463 0.302 1.53 0.13
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 371.38 on 301 degrees of freedom
## AIC: 375.4
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.234 0.153 -8.06 0.00000000000000078 ***
## orghave_incentivesYes -1.145 0.492 -2.33 0.02 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 294.68 on 301 degrees of freedom
## AIC: 298.7
##
## Number of Fisher Scoring iterations: 5Q11 Pred: None
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.121 0.132 0.92 0.36
## orghave_neitherYes -2.503 0.446 -5.61 0.000000021 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 361.90 on 301 degrees of freedom
## AIC: 365.9
##
## Number of Fisher Scoring iterations: 5Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.860 0.144 -5.99 0.0000000022 ***
## orghave_neitherYes -1.204 0.402 -3.00 0.0027 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 332.41 on 301 degrees of freedom
## AIC: 336.4
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.661 0.139 -4.77 0.0000018 ***
## orghave_neitherYes -0.932 0.346 -2.70 0.007 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 362.11 on 301 degrees of freedom
## AIC: 366.1
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.0345 0.1313 -0.26 0.79
## orghave_neitherYes -0.3945 0.2761 -1.43 0.15
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 416.79 on 301 degrees of freedom
## AIC: 420.8
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.313 0.133 -2.35 0.019 *
## orghave_neitherYes -0.767 0.303 -2.53 0.011 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 396.41 on 301 degrees of freedom
## AIC: 400.4
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.569 0.174 -9.03 <0.0000000000000002 ***
## orghave_neitherYes -1.552 0.615 -2.52 0.012 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 238.15 on 301 degrees of freedom
## AIC: 242.2
##
## Number of Fisher Scoring iterations: 5Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.585 0.137 -4.27 0.000019 ***
## orghave_neitherYes -1.345 0.382 -3.52 0.00043 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 356.57 on 301 degrees of freedom
## AIC: 360.6
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.548 0.136 -4.02 0.000058 ***
## orghave_neitherYes -1.516 0.399 -3.80 0.00015 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 354.84 on 301 degrees of freedom
## AIC: 358.8
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.869 0.193 -9.69 < 0.0000000000000002 ***
## orghave_neitherYes 1.499 0.309 4.85 0.0000012 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 278.48 on 301 degrees of freedom
## AIC: 282.5
##
## Number of Fisher Scoring iterations: 4Pred Q12 Challenges Clarity
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.255 0.239 -1.07 0.29
## challenges_clarityYes -0.058 0.274 -0.21 0.83
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 413.29 on 301 degrees of freedom
## AIC: 417.3
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.317 0.291 -4.53 0.0000059 ***
## challenges_clarityYes 0.309 0.326 0.95 0.34
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 342.57 on 301 degrees of freedom
## AIC: 346.6
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.696 0.328 -5.17 0.00000023 ***
## challenges_clarityYes 1.055 0.356 2.96 0.003 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 360.13 on 301 degrees of freedom
## AIC: 364.1
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.255 0.239 -1.07 0.29
## challenges_clarityYes 0.169 0.273 0.62 0.54
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 418.47 on 301 degrees of freedom
## AIC: 422.5
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.936 0.264 -3.55 0.00039 ***
## challenges_clarityYes 0.588 0.296 1.99 0.04676 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 399.11 on 301 degrees of freedom
## AIC: 403.1
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.064 0.375 -5.5 0.000000038 ***
## challenges_clarityYes 0.336 0.418 0.8 0.42
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 246.82 on 301 degrees of freedom
## AIC: 250.8
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.007 0.268 -3.76 0.00017 ***
## challenges_clarityYes 0.228 0.303 0.75 0.45115
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 371.45 on 301 degrees of freedom
## AIC: 375.5
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.235 0.284 -4.35 0.000014 ***
## challenges_clarityYes 0.535 0.316 1.69 0.091 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 370.65 on 301 degrees of freedom
## AIC: 374.6
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_clarity, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.495 0.307 -4.87 0.0000011 ***
## challenges_clarityYes 0.125 0.348 0.36 0.72
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 301.44 on 301 degrees of freedom
## AIC: 305.4
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of training
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.174 0.148 -1.18 0.24
## challenges_trainingYes -0.323 0.240 -1.35 0.18
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 411.52 on 301 degrees of freedom
## AIC: 415.5
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.879 0.162 -5.43 0.000000057 ***
## challenges_trainingYes -0.550 0.283 -1.94 0.052 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 339.57 on 301 degrees of freedom
## AIC: 343.6
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.653 0.155 -4.20 0.000027 ***
## challenges_trainingYes -0.526 0.266 -1.98 0.048 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 366.33 on 301 degrees of freedom
## AIC: 370.3
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.0435 0.1475 -0.29 0.77
## challenges_trainingYes -0.2100 0.2364 -0.89 0.37
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 418.06 on 301 degrees of freedom
## AIC: 422.1
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.396 0.150 -2.64 0.0084 **
## challenges_trainingYes -0.210 0.244 -0.86 0.3895
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 402.51 on 301 degrees of freedom
## AIC: 406.5
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.8045 0.2116 -8.53 <0.0000000000000002 ***
## challenges_trainingYes 0.0127 0.3368 0.04 0.97
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.5 on 302 degrees of freedom
## Residual deviance: 247.5 on 301 degrees of freedom
## AIC: 251.5
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.905 0.163 -5.56 0.000000027 ***
## challenges_trainingYes 0.186 0.254 0.73 0.46
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 371.49 on 301 degrees of freedom
## AIC: 375.5
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.8525 0.1610 -5.29 0.00000012 ***
## challenges_trainingYes 0.0956 0.2542 0.38 0.71
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 373.53 on 301 degrees of freedom
## AIC: 377.5
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_training, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.596 0.197 -8.11 0.00000000000000053 ***
## challenges_trainingYes 0.464 0.290 1.60 0.11
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 299.04 on 301 degrees of freedom
## AIC: 303
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of resources or tools
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.385 0.160 -2.41 0.016 *
## challenges_resourcesYes 0.184 0.233 0.79 0.429
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 412.71 on 301 degrees of freedom
## AIC: 416.7
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.449 0.200 -7.26 0.00000000000039 ***
## challenges_resourcesYes 0.734 0.269 2.73 0.0063 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 335.88 on 301 degrees of freedom
## AIC: 339.9
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.8737 0.1718 -5.08 0.00000037 ***
## challenges_resourcesYes 0.0602 0.2512 0.24 0.81
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 370.30 on 301 degrees of freedom
## AIC: 374.3
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.259 0.158 -1.64 0.10
## challenges_resourcesYes 0.288 0.231 1.24 0.21
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 417.31 on 301 degrees of freedom
## AIC: 421.3
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.647 0.165 -3.93 0.000086 ***
## challenges_resourcesYes 0.360 0.237 1.52 0.13
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 400.95 on 301 degrees of freedom
## AIC: 405
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.967 0.239 -8.24 <0.0000000000000002 ***
## challenges_resourcesYes 0.340 0.330 1.03 0.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 246.44 on 301 degrees of freedom
## AIC: 250.4
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.8154 0.1699 -4.80 0.0000016 ***
## challenges_resourcesYes -0.0319 0.2507 -0.13 0.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 372.01 on 301 degrees of freedom
## AIC: 376
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.026 0.178 -5.77 0.0000000077 ***
## challenges_resourcesYes 0.439 0.250 1.75 0.08 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 370.59 on 301 degrees of freedom
## AIC: 374.6
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_resources, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.489 0.202 -7.37 0.00000000000017 ***
## challenges_resourcesYes 0.190 0.289 0.66 0.51
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 301.14 on 301 degrees of freedom
## AIC: 305.1
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of incentives
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.349 0.147 -2.38 0.018 *
## challenges_incentivesYes 0.134 0.240 0.56 0.577
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 413.03 on 301 degrees of freedom
## AIC: 417
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.0916 0.1668 -6.54 0.00000000006 ***
## challenges_incentivesYes 0.0401 0.2727 0.15 0.88
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 343.47 on 301 degrees of freedom
## AIC: 347.5
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.120 0.168 -6.67 0.000000000026 ***
## challenges_incentivesYes 0.684 0.256 2.67 0.0076 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 363.23 on 301 degrees of freedom
## AIC: 367.2
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.178 0.145 -1.23 0.22
## challenges_incentivesYes 0.143 0.238 0.60 0.55
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 418.50 on 301 degrees of freedom
## AIC: 422.5
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.570 0.151 -3.78 0.00015 ***
## challenges_incentivesYes 0.246 0.244 1.01 0.31325
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 402.24 on 301 degrees of freedom
## AIC: 406.2
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.039 0.227 -9.00 <0.0000000000000002 ***
## challenges_incentivesYes 0.573 0.332 1.73 0.084 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 244.55 on 301 degrees of freedom
## AIC: 248.6
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.010 0.164 -6.17 0.00000000067 ***
## challenges_incentivesYes 0.461 0.255 1.80 0.071 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 368.79 on 301 degrees of freedom
## AIC: 372.8
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.037 0.165 -6.30 0.0000000003 ***
## challenges_incentivesYes 0.564 0.255 2.21 0.027 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 368.79 on 301 degrees of freedom
## AIC: 372.8
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.120 0.168 -6.67 0.000000000026 ***
## challenges_incentivesYes -0.911 0.339 -2.68 0.0073 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 293.57 on 301 degrees of freedom
## AIC: 297.6
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of understanding whether / why it may be valuable
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.278 0.131 -2.13 0.033 *
## challenges_valuableYes -0.102 0.286 -0.36 0.723
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 413.21 on 301 degrees of freedom
## AIC: 417.2
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.049 0.148 -7.11 0.0000000000012 ***
## challenges_valuableYes -0.135 0.330 -0.41 0.68
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 343.33 on 301 degrees of freedom
## AIC: 347.3
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.922 0.143 -6.43 0.00000000013 ***
## challenges_valuableYes 0.344 0.297 1.16 0.25
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 369.04 on 301 degrees of freedom
## AIC: 373
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.193 0.130 -1.49 0.14
## challenges_valuableYes 0.318 0.282 1.13 0.26
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 417.58 on 301 degrees of freedom
## AIC: 421.6
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.4864 0.1332 -3.65 0.00026 ***
## challenges_valuableYes 0.0417 0.2888 0.14 0.88527
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 403.23 on 301 degrees of freedom
## AIC: 407.2
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.831 0.188 -9.77 <0.0000000000000002 ***
## challenges_valuableYes 0.145 0.392 0.37 0.71
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 247.37 on 301 degrees of freedom
## AIC: 251.4
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.964 0.145 -6.66 0.000000000027 ***
## challenges_valuableYes 0.584 0.293 2.00 0.046 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 368.14 on 301 degrees of freedom
## AIC: 372.1
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.922 0.143 -6.43 0.00000000013 ***
## challenges_valuableYes 0.477 0.294 1.63 0.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 371.08 on 301 degrees of freedom
## AIC: 375.1
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_valuable, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.355 0.160 -8.46 <0.0000000000000002 ***
## challenges_valuableYes -0.217 0.368 -0.59 0.56
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 301.21 on 301 degrees of freedom
## AIC: 305.2
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of incentives
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.349 0.147 -2.38 0.018 *
## challenges_incentivesYes 0.134 0.240 0.56 0.577
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 413.03 on 301 degrees of freedom
## AIC: 417
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.0916 0.1668 -6.54 0.00000000006 ***
## challenges_incentivesYes 0.0401 0.2727 0.15 0.88
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 343.47 on 301 degrees of freedom
## AIC: 347.5
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.120 0.168 -6.67 0.000000000026 ***
## challenges_incentivesYes 0.684 0.256 2.67 0.0076 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 363.23 on 301 degrees of freedom
## AIC: 367.2
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.178 0.145 -1.23 0.22
## challenges_incentivesYes 0.143 0.238 0.60 0.55
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 418.50 on 301 degrees of freedom
## AIC: 422.5
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.570 0.151 -3.78 0.00015 ***
## challenges_incentivesYes 0.246 0.244 1.01 0.31325
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 402.24 on 301 degrees of freedom
## AIC: 406.2
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.039 0.227 -9.00 <0.0000000000000002 ***
## challenges_incentivesYes 0.573 0.332 1.73 0.084 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 244.55 on 301 degrees of freedom
## AIC: 248.6
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.010 0.164 -6.17 0.00000000067 ***
## challenges_incentivesYes 0.461 0.255 1.80 0.071 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 368.79 on 301 degrees of freedom
## AIC: 372.8
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.037 0.165 -6.30 0.0000000003 ***
## challenges_incentivesYes 0.564 0.255 2.21 0.027 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 368.79 on 301 degrees of freedom
## AIC: 372.8
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.120 0.168 -6.67 0.000000000026 ***
## challenges_incentivesYes -0.911 0.339 -2.68 0.0073 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 293.57 on 301 degrees of freedom
## AIC: 297.6
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of support
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.422 0.129 -3.26 0.0011 **
## challenges_supportYes 0.688 0.306 2.25 0.0245 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 408.22 on 301 degrees of freedom
## AIC: 412.2
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.0880 0.1457 -7.47 0.000000000000081 ***
## challenges_supportYes 0.0635 0.3439 0.18 0.85
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 343.46 on 301 degrees of freedom
## AIC: 347.5
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.944 0.141 -6.71 0.00000000002 ***
## challenges_supportYes 0.523 0.314 1.67 0.096 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 367.65 on 301 degrees of freedom
## AIC: 371.7
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.241 0.127 -1.89 0.058 .
## challenges_supportYes 0.662 0.308 2.15 0.032 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 414.14 on 301 degrees of freedom
## AIC: 418.1
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.422 0.129 -3.26 0.0011 **
## challenges_supportYes -0.328 0.321 -1.02 0.3073
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 402.19 on 301 degrees of freedom
## AIC: 406.2
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.992 0.195 -10.24 <0.0000000000000002 ***
## challenges_supportYes 0.869 0.374 2.32 0.02 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 242.51 on 301 degrees of freedom
## AIC: 246.5
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.905 0.140 -6.48 0.000000000091 ***
## challenges_supportYes 0.404 0.316 1.28 0.2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 370.43 on 301 degrees of freedom
## AIC: 374.4
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.905 0.140 -6.48 0.000000000091 ***
## challenges_supportYes 0.484 0.314 1.54 0.12
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 371.35 on 301 degrees of freedom
## AIC: 375.3
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_support, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.337 0.156 -8.58 <0.0000000000000002 ***
## challenges_supportYes -0.390 0.414 -0.94 0.35
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 300.63 on 301 degrees of freedom
## AIC: 304.6
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of clarity about expectations
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.396 0.129 -3.07 0.0021 **
## challenges_clarityexpectationsYes 0.550 0.307 1.79 0.0729 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 410.11 on 301 degrees of freedom
## AIC: 414.1
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.031 0.143 -7.19 0.00000000000065 ***
## challenges_clarityexpectationsYes -0.285 0.369 -0.77 0.44
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 342.87 on 301 degrees of freedom
## AIC: 346.9
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.990 0.142 -6.97 0.0000000000032 ***
## challenges_clarityexpectationsYes 0.758 0.313 2.42 0.016 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 364.65 on 301 degrees of freedom
## AIC: 368.6
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.264 0.127 -2.08 0.0378 *
## challenges_clarityexpectationsYes 0.817 0.315 2.59 0.0095 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 411.88 on 301 degrees of freedom
## AIC: 415.9
##
## Number of Fisher Scoring iterations: 4Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.530 0.131 -4.06 0.00005 ***
## challenges_clarityexpectationsYes 0.298 0.308 0.97 0.33
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 402.33 on 301 degrees of freedom
## AIC: 406.3
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.755 0.178 -9.86 <0.0000000000000002 ***
## challenges_clarityexpectationsYes -0.282 0.469 -0.60 0.55
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 247.12 on 301 degrees of freedom
## AIC: 251.1
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.853 0.138 -6.19 0.00000000062 ***
## challenges_clarityexpectationsYes 0.131 0.326 0.40 0.69
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 371.87 on 301 degrees of freedom
## AIC: 375.9
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.853 0.138 -6.19 0.00000000062 ***
## challenges_clarityexpectationsYes 0.217 0.322 0.67 0.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 373.23 on 301 degrees of freedom
## AIC: 377.2
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_clarityexpectations,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.294 0.154 -8.43 <0.0000000000000002 ***
## challenges_clarityexpectationsYes -0.743 0.460 -1.61 0.11
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 298.58 on 301 degrees of freedom
## AIC: 302.6
##
## Number of Fisher Scoring iterations: 4Pred: Q12 Lack of Trust
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.3087 0.1358 -2.27 0.023 *
## challenges_trustYes 0.0354 0.2622 0.14 0.892
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 413.32 on 301 degrees of freedom
## AIC: 417.3
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.135 0.156 -7.26 0.0000000000004 ***
## challenges_trustYes 0.210 0.292 0.72 0.47
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 342.98 on 301 degrees of freedom
## AIC: 347
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.039 0.153 -6.80 0.00000000001 ***
## challenges_trustYes 0.665 0.273 2.44 0.015 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 364.52 on 301 degrees of freedom
## AIC: 368.5
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.235 0.135 -1.74 0.082 .
## challenges_trustYes 0.409 0.261 1.57 0.117
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 416.39 on 301 degrees of freedom
## AIC: 420.4
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.593 0.140 -4.23 0.000023 ***
## challenges_trustYes 0.420 0.263 1.59 0.11
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 400.73 on 301 degrees of freedom
## AIC: 404.7
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.064 0.212 -9.72 <0.0000000000000002 ***
## challenges_trustYes 0.812 0.341 2.38 0.017 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 242.08 on 301 degrees of freedom
## AIC: 246.1
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.971 0.150 -6.46 0.00000000011 ***
## challenges_trustYes 0.492 0.274 1.80 0.072 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 368.84 on 301 degrees of freedom
## AIC: 372.8
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.971 0.150 -6.46 0.00000000011 ***
## challenges_trustYes 0.544 0.273 2.00 0.046 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 369.75 on 301 degrees of freedom
## AIC: 373.7
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_trust, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.235 0.161 -7.69 0.000000000000015 ***
## challenges_trustYes -0.725 0.374 -1.94 0.053 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 297.40 on 301 degrees of freedom
## AIC: 301.4
##
## Number of Fisher Scoring iterations: 4Pred: Q12 None
Take ethical / responsible AI trainings
##
## Call:
## glm(formula = q9_takeethical ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.349 0.137 -2.55 0.011 *
## challenges_noneYes 0.180 0.259 0.69 0.488
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 413.34 on 302 degrees of freedom
## Residual deviance: 412.86 on 301 degrees of freedom
## AIC: 416.9
##
## Number of Fisher Scoring iterations: 4Conduct audits of the gen AI tools
##
## Call:
## glm(formula = q9_conductaudits ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.224 0.161 -7.61 0.000000000000028 ***
## challenges_noneYes 0.494 0.284 1.74 0.082 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 343.49 on 302 degrees of freedom
## Residual deviance: 340.54 on 301 degrees of freedom
## AIC: 344.5
##
## Number of Fisher Scoring iterations: 4Conduct fairness / bias testing
##
## Call:
## glm(formula = q9_conductfair ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.84730 0.14712 -5.76 0.0000000085 ***
## challenges_noneYes 0.00573 0.28087 0.02 0.98
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 370.35 on 302 degrees of freedom
## Residual deviance: 370.35 on 301 degrees of freedom
## AIC: 374.4
##
## Number of Fisher Scoring iterations: 4Consider data privacy implications and take actions to protect data privacy
##
## Call:
## glm(formula = q9_considerdatapriv ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.091 0.135 -0.67 0.50
## challenges_noneYes -0.127 0.259 -0.49 0.62
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.85 on 302 degrees of freedom
## Residual deviance: 418.61 on 301 degrees of freedom
## AIC: 422.6
##
## Number of Fisher Scoring iterations: 3Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## glm(formula = q9_askaboutdata ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.4626 0.1385 -3.34 0.00083 ***
## challenges_noneYes -0.0546 0.2658 -0.21 0.83716
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.25 on 302 degrees of freedom
## Residual deviance: 403.21 on 301 degrees of freedom
## AIC: 407.2
##
## Number of Fisher Scoring iterations: 4Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## glm(formula = q9_adversarial ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.735 0.189 -9.19 <0.0000000000000002 ***
## challenges_noneYes -0.253 0.386 -0.66 0.51
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 247.50 on 302 degrees of freedom
## Residual deviance: 247.06 on 301 degrees of freedom
## AIC: 251.1
##
## Number of Fisher Scoring iterations: 4Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## glm(formula = q9_explainability ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.762 0.145 -5.27 0.00000014 ***
## challenges_noneYes -0.258 0.288 -0.90 0.37
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 372.03 on 302 degrees of freedom
## Residual deviance: 371.21 on 301 degrees of freedom
## AIC: 375.2
##
## Number of Fisher Scoring iterations: 4Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## glm(formula = q9_buildtransparency ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.8044 0.1459 -5.51 0.000000035 ***
## challenges_noneYes -0.0372 0.2802 -0.13 0.89
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 373.67 on 302 degrees of freedom
## Residual deviance: 373.66 on 301 degrees of freedom
## AIC: 377.7
##
## Number of Fisher Scoring iterations: 4No actions taken (to my knowledge)
##
## Call:
## glm(formula = q9_noactions ~ challenges_none, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.566 0.178 -8.78 <0.0000000000000002 ***
## challenges_noneYes 0.547 0.306 1.79 0.074 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 301.57 on 302 degrees of freedom
## Residual deviance: 298.48 on 301 degrees of freedom
## AIC: 302.5
##
## Number of Fisher Scoring iterations: 4Q12: What types of challenges have you or your team faced in regards to using gen AI responsibly?
Q11 Pred: Leadership that has expressed commitment to responsible AI
Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.114 0.174 6.41 0.00000000015 ***
## orghave_leadershipYes 0.175 0.278 0.63 0.53
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 329.94 on 302 degrees of freedom
## Residual deviance: 329.53 on 301 degrees of freedom
## AIC: 333.5
##
## Number of Fisher Scoring iterations: 4Lack of training
##
## Call:
## glm(formula = challenges_training ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.203 0.151 -1.35 0.178
## orghave_leadershipYes -0.588 0.245 -2.40 0.016 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 405.99 on 302 degrees of freedom
## Residual deviance: 400.11 on 301 degrees of freedom
## AIC: 404.1
##
## Number of Fisher Scoring iterations: 4Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.364 0.152 -2.39 0.017 *
## orghave_leadershipYes 0.508 0.235 2.16 0.031 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.30 on 302 degrees of freedom
## Residual deviance: 413.61 on 301 degrees of freedom
## AIC: 417.6
##
## Number of Fisher Scoring iterations: 4Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.5048 0.1547 -3.26 0.0011 **
## orghave_leadershipYes -0.0705 0.2422 -0.29 0.7709
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 399.21 on 302 degrees of freedom
## Residual deviance: 399.12 on 301 degrees of freedom
## AIC: 403.1
##
## Number of Fisher Scoring iterations: 4Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.3378 0.1847 -7.24 0.00000000000044 ***
## orghave_leadershipYes 0.0487 0.2852 0.17 0.86
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 312.44 on 302 degrees of freedom
## Residual deviance: 312.41 on 301 degrees of freedom
## AIC: 316.4
##
## Number of Fisher Scoring iterations: 4Lack of support
##
## Call:
## glm(formula = challenges_support ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.407 0.189 -7.46 0.000000000000084 ***
## orghave_leadershipYes -0.375 0.317 -1.18 0.24
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.94 on 302 degrees of freedom
## Residual deviance: 279.51 on 301 degrees of freedom
## AIC: 283.5
##
## Number of Fisher Scoring iterations: 4Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ orghave_leadership,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.5960 0.2002 -7.97 0.0000000000000016 ***
## orghave_leadershipYes 0.0523 0.3086 0.17 0.87
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.82 on 302 degrees of freedom
## Residual deviance: 277.79 on 301 degrees of freedom
## AIC: 281.8
##
## Number of Fisher Scoring iterations: 3Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.9963 0.1689 -5.90 0.0000000037 ***
## orghave_leadershipYes -0.0289 0.2640 -0.11 0.91
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.83 on 302 degrees of freedom
## Residual deviance: 351.82 on 301 degrees of freedom
## AIC: 355.8
##
## Number of Fisher Scoring iterations: 4None
##
## Call:
## glm(formula = challenges_none ~ orghave_leadership, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.206 0.178 -6.78 0.000000000012 ***
## orghave_leadershipYes 0.525 0.260 2.02 0.043 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 355.80 on 302 degrees of freedom
## Residual deviance: 351.71 on 301 degrees of freedom
## AIC: 355.7
##
## Number of Fisher Scoring iterations: 4Pred: Responsible or ethical AI principles
Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.023 0.172 5.95 0.0000000026 ***
## orghave_responsibleYes 0.402 0.281 1.43 0.15
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 329.94 on 302 degrees of freedom
## Residual deviance: 327.85 on 301 degrees of freedom
## AIC: 331.9
##
## Number of Fisher Scoring iterations: 4Lack of training
##
## Call:
## glm(formula = challenges_training ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.208 0.152 -1.36 0.173
## orghave_responsibleYes -0.556 0.243 -2.29 0.022 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 405.99 on 302 degrees of freedom
## Residual deviance: 400.66 on 301 degrees of freedom
## AIC: 404.7
##
## Number of Fisher Scoring iterations: 4Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.208 0.152 -1.36 0.17
## orghave_responsibleYes 0.130 0.233 0.56 0.58
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.30 on 302 degrees of freedom
## Residual deviance: 417.99 on 301 degrees of freedom
## AIC: 422
##
## Number of Fisher Scoring iterations: 3Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.693 0.161 -4.31 0.000016 ***
## orghave_responsibleYes 0.365 0.240 1.52 0.13
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 399.21 on 302 degrees of freedom
## Residual deviance: 396.91 on 301 degrees of freedom
## AIC: 400.9
##
## Number of Fisher Scoring iterations: 4Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.209 0.180 -6.71 0.000000000019 ***
## orghave_responsibleYes -0.267 0.289 -0.92 0.36
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 312.44 on 302 degrees of freedom
## Residual deviance: 311.57 on 301 degrees of freedom
## AIC: 315.6
##
## Number of Fisher Scoring iterations: 4Lack of support
##
## Call:
## glm(formula = challenges_support ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.415 0.191 -7.40 0.00000000000013 ***
## orghave_responsibleYes -0.341 0.313 -1.09 0.28
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.94 on 302 degrees of freedom
## Residual deviance: 279.74 on 301 degrees of freedom
## AIC: 283.7
##
## Number of Fisher Scoring iterations: 4Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ orghave_responsible,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.6094 0.2034 -7.91 0.0000000000000025 ***
## orghave_responsibleYes 0.0815 0.3071 0.27 0.79
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.82 on 302 degrees of freedom
## Residual deviance: 277.75 on 301 degrees of freedom
## AIC: 281.7
##
## Number of Fisher Scoring iterations: 4Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.177 0.179 -6.59 0.000000000045 ***
## orghave_responsibleYes 0.377 0.261 1.44 0.15
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.83 on 302 degrees of freedom
## Residual deviance: 349.75 on 301 degrees of freedom
## AIC: 353.7
##
## Number of Fisher Scoring iterations: 4None
##
## Call:
## glm(formula = challenges_none ~ orghave_responsible, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.083 0.174 -6.21 0.00000000053 ***
## orghave_responsibleYes 0.247 0.259 0.95 0.34
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 355.80 on 302 degrees of freedom
## Residual deviance: 354.89 on 301 degrees of freedom
## AIC: 358.9
##
## Number of Fisher Scoring iterations: 4Q11 Pred: A policy/policies that inform the use of gen AI
Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.050 0.166 6.32 0.00000000025 ***
## orghave_policyYes 0.381 0.290 1.32 0.19
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 329.94 on 302 degrees of freedom
## Residual deviance: 328.16 on 301 degrees of freedom
## AIC: 332.2
##
## Number of Fisher Scoring iterations: 4Lack of training
##
## Call:
## glm(formula = challenges_training ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.245 0.147 -1.67 0.095 .
## orghave_policyYes -0.529 0.249 -2.12 0.034 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 405.99 on 302 degrees of freedom
## Residual deviance: 401.40 on 301 degrees of freedom
## AIC: 405.4
##
## Number of Fisher Scoring iterations: 4Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.202 0.146 -1.38 0.17
## orghave_policyYes 0.132 0.238 0.55 0.58
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.30 on 302 degrees of freedom
## Residual deviance: 417.99 on 301 degrees of freedom
## AIC: 422
##
## Number of Fisher Scoring iterations: 3Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.741 0.156 -4.76 0.0000019 ***
## orghave_policyYes 0.530 0.244 2.17 0.03 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 399.21 on 302 degrees of freedom
## Residual deviance: 394.51 on 301 degrees of freedom
## AIC: 398.5
##
## Number of Fisher Scoring iterations: 4Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.31507 0.17808 -7.38 0.00000000000015 ***
## orghave_policyYes -0.00669 0.29067 -0.02 0.98
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 312.44 on 302 degrees of freedom
## Residual deviance: 312.43 on 301 degrees of freedom
## AIC: 316.4
##
## Number of Fisher Scoring iterations: 4Lack of support
##
## Call:
## glm(formula = challenges_support ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.591 0.194 -8.20 0.00000000000000024 ***
## orghave_policyYes 0.102 0.310 0.33 0.74
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.94 on 302 degrees of freedom
## Residual deviance: 280.83 on 301 degrees of freedom
## AIC: 284.8
##
## Number of Fisher Scoring iterations: 4Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ orghave_policy,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.668 0.199 -8.38 <0.0000000000000002 ***
## orghave_policyYes 0.237 0.310 0.76 0.44
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.82 on 302 degrees of freedom
## Residual deviance: 277.24 on 301 degrees of freedom
## AIC: 281.2
##
## Number of Fisher Scoring iterations: 4Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.106 0.168 -6.57 0.00000000005 ***
## orghave_policyYes 0.250 0.265 0.94 0.35
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.83 on 302 degrees of freedom
## Residual deviance: 350.95 on 301 degrees of freedom
## AIC: 354.9
##
## Number of Fisher Scoring iterations: 4None
##
## Call:
## glm(formula = challenges_none ~ orghave_policy, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.078 0.167 -6.45 0.00000000011 ***
## orghave_policyYes 0.263 0.263 1.00 0.32
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 355.8 on 302 degrees of freedom
## Residual deviance: 354.8 on 301 degrees of freedom
## AIC: 358.8
##
## Number of Fisher Scoring iterations: 4Q11 Pred: Clear incentives for using/implementing gen AI responsibly
Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.121 0.149 7.54 0.000000000000048 ***
## orghave_incentivesYes 0.353 0.366 0.96 0.33
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 329.94 on 302 degrees of freedom
## Residual deviance: 328.96 on 301 degrees of freedom
## AIC: 333
##
## Number of Fisher Scoring iterations: 4Lack of training
##
## Call:
## glm(formula = challenges_training ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.365 0.130 -2.80 0.0051 **
## orghave_incentivesYes -0.380 0.308 -1.23 0.2168
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 405.99 on 302 degrees of freedom
## Residual deviance: 404.43 on 301 degrees of freedom
## AIC: 408.4
##
## Number of Fisher Scoring iterations: 4Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.181 0.129 -1.41 0.16
## orghave_incentivesYes 0.147 0.290 0.51 0.61
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.30 on 302 degrees of freedom
## Residual deviance: 418.04 on 301 degrees of freedom
## AIC: 422
##
## Number of Fisher Scoring iterations: 3Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.573 0.133 -4.29 0.000018 ***
## orghave_incentivesYes 0.195 0.297 0.66 0.51
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 399.21 on 302 degrees of freedom
## Residual deviance: 398.78 on 301 degrees of freedom
## AIC: 402.8
##
## Number of Fisher Scoring iterations: 4Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.356 0.159 -8.55 <0.0000000000000002 ***
## orghave_incentivesYes 0.188 0.345 0.55 0.58
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 312.44 on 302 degrees of freedom
## Residual deviance: 312.14 on 301 degrees of freedom
## AIC: 316.1
##
## Number of Fisher Scoring iterations: 4Lack of support
##
## Call:
## glm(formula = challenges_support ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.514 0.167 -9.09 <0.0000000000000002 ***
## orghave_incentivesYes -0.201 0.399 -0.50 0.61
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.94 on 302 degrees of freedom
## Residual deviance: 280.68 on 301 degrees of freedom
## AIC: 284.7
##
## Number of Fisher Scoring iterations: 4Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ orghave_incentives,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.5706 0.1696 -9.26 <0.0000000000000002 ***
## orghave_incentivesYes -0.0186 0.3862 -0.05 0.96
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.82 on 302 degrees of freedom
## Residual deviance: 277.81 on 301 degrees of freedom
## AIC: 281.8
##
## Number of Fisher Scoring iterations: 3Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.143 0.150 -7.64 0.000000000000021 ***
## orghave_incentivesYes 0.623 0.308 2.02 0.043 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.83 on 302 degrees of freedom
## Residual deviance: 347.87 on 301 degrees of freedom
## AIC: 351.9
##
## Number of Fisher Scoring iterations: 4None
##
## Call:
## glm(formula = challenges_none ~ orghave_incentives, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.034 0.146 -7.11 0.0000000000012 ***
## orghave_incentivesYes 0.290 0.314 0.92 0.36
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 355.80 on 302 degrees of freedom
## Residual deviance: 354.97 on 301 degrees of freedom
## AIC: 359
##
## Number of Fisher Scoring iterations: 4Q11 Pred: None
Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.1688 0.1544 7.57 0.000000000000037 ***
## orghave_neitherYes 0.0659 0.3233 0.20 0.84
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 329.94 on 302 degrees of freedom
## Residual deviance: 329.89 on 301 degrees of freedom
## AIC: 333.9
##
## Number of Fisher Scoring iterations: 4Lack of training
##
## Call:
## glm(formula = challenges_training ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.511 0.136 -3.77 0.00017 ***
## orghave_neitherYes 0.313 0.274 1.14 0.25396
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 405.99 on 302 degrees of freedom
## Residual deviance: 404.70 on 301 degrees of freedom
## AIC: 408.7
##
## Number of Fisher Scoring iterations: 4Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.104 0.131 -0.79 0.43
## orghave_neitherYes -0.209 0.274 -0.76 0.45
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 418.30 on 302 degrees of freedom
## Residual deviance: 417.72 on 301 degrees of freedom
## AIC: 421.7
##
## Number of Fisher Scoring iterations: 3Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.5478 0.1363 -4.02 0.000058 ***
## orghave_neitherYes 0.0594 0.2799 0.21 0.83
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 399.21 on 302 degrees of freedom
## Residual deviance: 399.16 on 301 degrees of freedom
## AIC: 403.2
##
## Number of Fisher Scoring iterations: 4Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.397 0.165 -8.48 <0.0000000000000002 ***
## orghave_neitherYes 0.317 0.319 1.00 0.32
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 312.44 on 302 degrees of freedom
## Residual deviance: 311.47 on 301 degrees of freedom
## AIC: 315.5
##
## Number of Fisher Scoring iterations: 4Lack of support
##
## Call:
## glm(formula = challenges_support ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.630 0.177 -9.19 <0.0000000000000002 ***
## orghave_neitherYes 0.313 0.341 0.92 0.36
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.94 on 302 degrees of freedom
## Residual deviance: 280.12 on 301 degrees of freedom
## AIC: 284.1
##
## Number of Fisher Scoring iterations: 4Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ orghave_neither,
## family = "binomial", data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.481 0.169 -8.76 <0.0000000000000002 ***
## orghave_neitherYes -0.449 0.395 -1.14 0.25
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.82 on 302 degrees of freedom
## Residual deviance: 276.43 on 301 degrees of freedom
## AIC: 280.4
##
## Number of Fisher Scoring iterations: 4Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.076 0.151 -7.14 0.00000000000097 ***
## orghave_neitherYes 0.275 0.298 0.92 0.36
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.83 on 302 degrees of freedom
## Residual deviance: 350.99 on 301 degrees of freedom
## AIC: 355
##
## Number of Fisher Scoring iterations: 4None
##
## Call:
## glm(formula = challenges_none ~ orghave_neither, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.944 0.146 -6.45 0.00000000011 ***
## orghave_neitherYes -0.136 0.310 -0.44 0.66
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 355.8 on 302 degrees of freedom
## Residual deviance: 355.6 on 301 degrees of freedom
## AIC: 359.6
##
## Number of Fisher Scoring iterations: 4Q13: On a scale of 1 to 5, to what extent do you agree that the following are challenges to using AI responsibly?
Q11 Pred: Leadership that has expressed commitment to responsible AI
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ orghave_leadership, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.680 -0.627 0.320 0.373 1.373
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6271 0.0838 43.30 <0.0000000000000002 ***
## orghave_leadershipYes 0.0529 0.1302 0.41 0.68
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00055, Adjusted R-squared: -0.00278
## F-statistic: 0.165 on 1 and 300 DF, p-value: 0.685Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ orghave_leadership, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.616 -0.616 -0.560 0.440 2.440
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6158 0.0826 31.65 <0.0000000000000002 ***
## orghave_leadershipYes -0.0558 0.1285 -0.43 0.66
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000629, Adjusted R-squared: -0.0027
## F-statistic: 0.189 on 1 and 300 DF, p-value: 0.664Discomfort raising issues around responsibility
##
## Call:
## lm(formula = Q13_3 ~ orghave_leadership, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.9545 -0.9545 0.0455 1.0455 2.1360
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9545 0.0863 34.22 <0.0000000000000002 ***
## orghave_leadershipYes -0.0905 0.1340 -0.68 0.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.15 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00153, Adjusted R-squared: -0.00181
## F-statistic: 0.457 on 1 and 299 DF, p-value: 0.5Other
##
## Call:
## lm(formula = Q13_4 ~ orghave_leadership, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.772 -0.772 0.228 0.312 2.312
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7724 0.1067 25.99 <0.0000000000000002 ***
## orghave_leadershipYes -0.0842 0.1626 -0.52 0.61
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00125, Adjusted R-squared: -0.00342
## F-statistic: 0.268 on 1 and 214 DF, p-value: 0.605Q11 Pred: Responsible or ethical AI principles
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ orghave_responsible, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.690 -0.619 0.310 0.382 1.381
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6185 0.0847 42.72 <0.0000000000000002 ***
## orghave_responsibleYes 0.0714 0.1296 0.55 0.58
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00101, Adjusted R-squared: -0.00232
## F-statistic: 0.304 on 1 and 300 DF, p-value: 0.582Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ orghave_responsible, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.651 -0.651 -0.549 0.451 2.451
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5491 0.0835 30.5 <0.0000000000000002 ***
## orghave_responsibleYes 0.1020 0.1278 0.8 0.43
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00212, Adjusted R-squared: -0.00121
## F-statistic: 0.637 on 1 and 300 DF, p-value: 0.425Discomfort raising issues around responsibility
##
## Call:
## lm(formula = Q13_3 ~ orghave_responsible, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.046 -0.820 0.180 0.954 2.180
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.820 0.087 32.42 <0.0000000000000002 ***
## orghave_responsibleYes 0.227 0.133 1.71 0.089 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00965, Adjusted R-squared: 0.00633
## F-statistic: 2.91 on 1 and 299 DF, p-value: 0.089Other
##
## Call:
## lm(formula = Q13_4 ~ orghave_responsible, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.776 -0.775 0.297 0.297 2.297
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7034 0.1089 24.82 <0.0000000000000002 ***
## orghave_responsibleYes 0.0721 0.1617 0.45 0.66
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.000928, Adjusted R-squared: -0.00374
## F-statistic: 0.199 on 1 and 214 DF, p-value: 0.656Q11 Pred: A policy/policies that inform the use of gen AI
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ orghave_policy, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.649 -0.649 0.351 0.351 1.351
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.648936 0.081301 44.9 <0.0000000000000002 ***
## orghave_policyYes 0.000187 0.132326 0.0 1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 6.63e-09, Adjusted R-squared: -0.00333
## F-statistic: 1.99e-06 on 1 and 300 DF, p-value: 0.999Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ orghave_policy, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.633 -0.633 -0.526 0.474 2.474
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6330 0.0801 32.86 <0.0000000000000002 ***
## orghave_policyYes -0.1067 0.1304 -0.82 0.41
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00222, Adjusted R-squared: -0.0011
## F-statistic: 0.669 on 1 and 300 DF, p-value: 0.414Discomfort raising issues around responsibility
##
## Call:
## lm(formula = Q13_3 ~ orghave_policy, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.9893 -0.9893 0.0107 1.0107 2.2018
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9893 0.0836 35.78 <0.0000000000000002 ***
## orghave_policyYes -0.1911 0.1358 -1.41 0.16
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00658, Adjusted R-squared: 0.00326
## F-statistic: 1.98 on 1 and 299 DF, p-value: 0.16Other
##
## Call:
## lm(formula = Q13_4 ~ orghave_policy, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.837 -0.838 0.324 0.324 2.324
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.676 0.101 26.42 <0.0000000000000002 ***
## orghave_policyYes 0.161 0.166 0.97 0.33
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00435, Adjusted R-squared: -0.000298
## F-statistic: 0.936 on 1 and 214 DF, p-value: 0.334Q11 Pred: Clear incentives for using/implementing gen AI responsibly
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ orghave_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.667 -0.667 0.333 0.424 1.424
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6667 0.0715 51.30 <0.0000000000000002 ***
## orghave_incentivesYes -0.0904 0.1617 -0.56 0.58
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00104, Adjusted R-squared: -0.00229
## F-statistic: 0.312 on 1 and 300 DF, p-value: 0.577Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ orghave_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.597 -0.597 -0.576 0.424 2.424
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5967 0.0706 36.80 <0.0000000000000002 ***
## orghave_incentivesYes -0.0204 0.1596 -0.13 0.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 5.46e-05, Adjusted R-squared: -0.00328
## F-statistic: 0.0164 on 1 and 300 DF, p-value: 0.898Discomfort raising issues around responsibility
##
## Call:
## lm(formula = Q13_3 ~ orghave_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.017 -0.893 0.107 1.107 2.107
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8926 0.0736 39.29 <0.0000000000000002 ***
## orghave_incentivesYes 0.1244 0.1663 0.75 0.46
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.15 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00187, Adjusted R-squared: -0.00147
## F-statistic: 0.56 on 1 and 299 DF, p-value: 0.455Other
##
## Call:
## lm(formula = Q13_4 ~ orghave_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.064 -1.064 0.355 0.355 2.355
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6450 0.0901 29.36 <0.0000000000000002 ***
## orghave_incentivesYes 0.4189 0.1931 2.17 0.031 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.17 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.0215, Adjusted R-squared: 0.0169
## F-statistic: 4.7 on 1 and 214 DF, p-value: 0.0312Q11 Pred: None
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ orghave_neither, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.667 -0.667 0.333 0.408 1.409
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6667 0.0733 50.0 <0.0000000000000002 ***
## orghave_neitherYes -0.0751 0.1512 -0.5 0.62
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000822, Adjusted R-squared: -0.00251
## F-statistic: 0.247 on 1 and 300 DF, p-value: 0.62Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ orghave_neither, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.690 -0.563 -0.563 0.437 2.437
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5628 0.0723 35.46 <0.0000000000000002 ***
## orghave_neitherYes 0.1274 0.1491 0.85 0.39
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00243, Adjusted R-squared: -0.000897
## F-statistic: 0.73 on 1 and 300 DF, p-value: 0.394Discomfort raising issues around responsibility
##
## Call:
## lm(formula = Q13_3 ~ orghave_neither, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.028 -0.883 0.117 1.117 2.117
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8826 0.0755 38.19 <0.0000000000000002 ***
## orghave_neitherYes 0.1456 0.1554 0.94 0.35
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00293, Adjusted R-squared: -0.000409
## F-statistic: 0.877 on 1 and 299 DF, p-value: 0.35Other
##
## Call:
## lm(formula = Q13_4 ~ orghave_neither, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.911 -0.745 0.310 0.310 2.310
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6901 0.0903 29.80 <0.0000000000000002 ***
## orghave_neitherYes 0.2211 0.1978 1.12 0.26
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.0058, Adjusted R-squared: 0.00116
## F-statistic: 1.25 on 1 and 214 DF, p-value: 0.265Q12 Pred: Lack of clarity on what that looks like
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_clarity, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.698 -0.698 0.302 0.514 1.514
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.486 0.133 26.2 <0.0000000000000002 ***
## challenges_clarityYes 0.213 0.152 1.4 0.16
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00652, Adjusted R-squared: 0.00321
## F-statistic: 1.97 on 1 and 300 DF, p-value: 0.162Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_clarity, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.686 -0.565 -0.565 0.435 2.435
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.686 0.131 20.45 <0.0000000000000002 ***
## challenges_clarityYes -0.121 0.150 -0.81 0.42
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00217, Adjusted R-squared: -0.00115
## F-statistic: 0.653 on 1 and 300 DF, p-value: 0.42Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_clarity, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.9714 -0.9004 0.0996 1.0996 2.0996
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.971 0.137 21.69 <0.0000000000000002 ***
## challenges_clarityYes -0.071 0.156 -0.45 0.65
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.15 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.000689, Adjusted R-squared: -0.00265
## F-statistic: 0.206 on 1 and 299 DF, p-value: 0.65Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_clarity, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.755 -0.755 0.270 0.270 2.270
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7547 0.1626 16.94 <0.0000000000000002 ***
## challenges_clarityYes -0.0247 0.1872 -0.13 0.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 8.11e-05, Adjusted R-squared: -0.00459
## F-statistic: 0.0173 on 1 and 214 DF, p-value: 0.895Q12 Pred: Lack of training
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_training, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.706 -0.612 0.294 0.388 1.388
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6120 0.0823 43.87 <0.0000000000000002 ***
## challenges_trainingYes 0.0939 0.1312 0.72 0.47
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0017, Adjusted R-squared: -0.00162
## F-statistic: 0.512 on 1 and 300 DF, p-value: 0.475Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_training, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.731 -0.731 -0.503 0.497 2.497
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5027 0.0809 30.94 <0.0000000000000002 ***
## challenges_trainingYes 0.2284 0.1288 1.77 0.077 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.09 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0104, Adjusted R-squared: 0.00706
## F-statistic: 3.14 on 1 and 300 DF, p-value: 0.0774Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_training, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.161 -0.760 0.240 0.839 2.240
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7596 0.0835 33.05 <0.0000000000000002 ***
## challenges_trainingYes 0.4015 0.1333 3.01 0.0028 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.13 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0294, Adjusted R-squared: 0.0262
## F-statistic: 9.07 on 1 and 299 DF, p-value: 0.00283Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_training, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.887 -0.887 0.113 0.353 2.353
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.647 0.101 26.20 <0.0000000000000002 ***
## challenges_trainingYes 0.240 0.166 1.45 0.15
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00971, Adjusted R-squared: 0.00508
## F-statistic: 2.1 on 1 and 214 DF, p-value: 0.149Q12 Pred: Lack of resources or tools
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_resources, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.772 -0.507 0.228 0.493 1.493
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.772 0.087 43.37 <0.0000000000000002 ***
## challenges_resourcesYes -0.265 0.128 -2.07 0.039 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0141, Adjusted R-squared: 0.0108
## F-statistic: 4.29 on 1 and 300 DF, p-value: 0.0393Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_resources, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.617 -0.617 -0.564 0.436 2.436
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6173 0.0864 30.30 <0.0000000000000002 ***
## challenges_resourcesYes -0.0530 0.1269 -0.42 0.68
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000581, Adjusted R-squared: -0.00275
## F-statistic: 0.174 on 1 and 300 DF, p-value: 0.676Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_resources, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.0556 -1.0556 -0.0556 0.9444 2.2446
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.0556 0.0893 34.22 <0.0000000000000002 ***
## challenges_resourcesYes -0.3002 0.1314 -2.28 0.023 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0172, Adjusted R-squared: 0.0139
## F-statistic: 5.22 on 1 and 299 DF, p-value: 0.023Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_resources, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.832 -0.832 0.168 0.369 2.369
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.832 0.111 25.52 <0.0000000000000002 ***
## challenges_resourcesYes -0.201 0.161 -1.25 0.21
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00724, Adjusted R-squared: 0.0026
## F-statistic: 1.56 on 1 and 214 DF, p-value: 0.213Q12 Pred: Lack incentives
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.0268 -0.4263 -0.0268 0.5737 1.5737
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4263 0.0781 43.89 < 0.0000000000000002 ***
## challenges_incentivesYes 0.6005 0.1282 4.68 0.0000043 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.08 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0682, Adjusted R-squared: 0.065
## F-statistic: 21.9 on 1 and 300 DF, p-value: 0.00000427Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.696 -0.696 -0.532 0.468 2.468
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5316 0.0796 31.81 <0.0000000000000002 ***
## challenges_incentivesYes 0.1648 0.1307 1.26 0.21
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00528, Adjusted R-squared: 0.00196
## F-statistic: 1.59 on 1 and 300 DF, p-value: 0.208Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.134 -0.788 0.212 1.212 2.212
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7884 0.0825 33.80 <0.0000000000000002 ***
## challenges_incentivesYes 0.3456 0.1352 2.56 0.011 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.13 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0214, Adjusted R-squared: 0.0181
## F-statistic: 6.53 on 1 and 299 DF, p-value: 0.0111Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_incentives, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.026 -0.686 0.427 0.427 2.428
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.572 0.099 25.97 <0.0000000000000002 ***
## challenges_incentivesYes 0.453 0.165 2.75 0.0065 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.16 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.0341, Adjusted R-squared: 0.0296
## F-statistic: 7.56 on 1 and 214 DF, p-value: 0.00648Q12 Pred: Lack of understanding whether / why it may be valuable
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_valuable, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.563 -0.563 0.437 0.437 1.437
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.5630 0.0714 49.87 <0.0000000000000002 ***
## challenges_valuableYes 0.4057 0.1552 2.61 0.0094 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0223, Adjusted R-squared: 0.019
## F-statistic: 6.83 on 1 and 300 DF, p-value: 0.0094Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_valuable, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.766 -0.546 -0.546 0.454 2.454
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5462 0.0711 35.83 <0.0000000000000002 ***
## challenges_valuableYes 0.2194 0.1543 1.42 0.16
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00669, Adjusted R-squared: 0.00338
## F-statistic: 2.02 on 1 and 300 DF, p-value: 0.156Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_valuable, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.125 -0.861 0.139 1.139 2.139
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8608 0.0741 38.59 <0.0000000000000002 ***
## challenges_valuableYes 0.2642 0.1608 1.64 0.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00895, Adjusted R-squared: 0.00564
## F-statistic: 2.7 on 1 and 299 DF, p-value: 0.101Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_valuable, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.932 -0.686 0.314 0.314 2.314
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.686 0.090 29.86 <0.0000000000000002 ***
## challenges_valuableYes 0.246 0.199 1.23 0.22
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00706, Adjusted R-squared: 0.00242
## F-statistic: 1.52 on 1 and 214 DF, p-value: 0.219Q12 Pred: Lack of support
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_support, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.849 -0.606 0.394 0.394 1.394
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6064 0.0704 51.23 <0.0000000000000002 ***
## challenges_supportYes 0.2426 0.1680 1.44 0.15
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0069, Adjusted R-squared: 0.00359
## F-statistic: 2.08 on 1 and 300 DF, p-value: 0.15Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_support, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.849 -0.538 -0.538 0.462 2.462
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5382 0.0693 36.63 <0.0000000000000002 ***
## challenges_supportYes 0.3109 0.1654 1.88 0.061 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.09 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0116, Adjusted R-squared: 0.00834
## F-statistic: 3.53 on 1 and 300 DF, p-value: 0.0611Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_support, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.346 -0.827 0.173 1.173 2.173
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8273 0.0716 39.50 <0.0000000000000002 ***
## challenges_supportYes 0.5188 0.1722 3.01 0.0028 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.13 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0295, Adjusted R-squared: 0.0262
## F-statistic: 9.08 on 1 and 299 DF, p-value: 0.00281Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_valuable, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.932 -0.686 0.314 0.314 2.314
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.686 0.090 29.86 <0.0000000000000002 ***
## challenges_valuableYes 0.246 0.199 1.23 0.22
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00706, Adjusted R-squared: 0.00242
## F-statistic: 1.52 on 1 and 214 DF, p-value: 0.219Q12 Pred: Lack of clarity about expectations or relevance to my role
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_clarityexpectations, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.79 -0.62 0.38 0.38 1.38
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6200 0.0704 51.43 <0.0000000000000002 ***
## challenges_clarityexpectationsYes 0.1685 0.1696 0.99 0.32
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00328, Adjusted R-squared: -4.55e-05
## F-statistic: 0.986 on 1 and 300 DF, p-value: 0.321Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_clarityexpectations, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.865 -0.536 -0.536 0.464 2.464
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.5360 0.0691 36.69 <0.0000000000000002 ***
## challenges_clarityexpectationsYes 0.3294 0.1666 1.98 0.049 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.09 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0129, Adjusted R-squared: 0.00958
## F-statistic: 3.91 on 1 and 300 DF, p-value: 0.0489Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_clarityexpectations, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.231 -0.851 0.149 1.149 2.149
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8514 0.0721 39.56 <0.0000000000000002 ***
## challenges_clarityexpectationsYes 0.3794 0.1734 2.19 0.029 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0158, Adjusted R-squared: 0.0125
## F-statistic: 4.79 on 1 and 299 DF, p-value: 0.0295Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_clarityexpectations, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.000 -0.768 0.310 0.310 2.310
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6902 0.0869 30.96 <0.0000000000000002 ***
## challenges_clarityexpectationsYes 0.3098 0.2258 1.37 0.17
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.00872, Adjusted R-squared: 0.00409
## F-statistic: 1.88 on 1 and 214 DF, p-value: 0.171Q12 Pred: Lack of trust
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_trust, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.498 -0.498 0.502 0.502 1.502
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4977 0.0731 47.9 < 0.0000000000000002 ***
## challenges_trustYes 0.5640 0.1411 4.0 0.000081 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.09 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0506, Adjusted R-squared: 0.0474
## F-statistic: 16 on 1 and 300 DF, p-value: 0.0000806Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_trust, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.593 -0.593 -0.593 0.407 2.407
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.592760 0.073984 35 <0.0000000000000002 ***
## challenges_trustYes -0.000168 0.142857 0 1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.1 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 4.59e-09, Adjusted R-squared: -0.00333
## F-statistic: 1.38e-06 on 1 and 300 DF, p-value: 0.999Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_trust, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.160 -0.827 0.173 1.173 2.173
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8273 0.0766 36.89 <0.0000000000000002 ***
## challenges_trustYes 0.3332 0.1477 2.26 0.025 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.14 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0167, Adjusted R-squared: 0.0134
## F-statistic: 5.09 on 1 and 299 DF, p-value: 0.0248Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_trust, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.051 -0.618 0.382 0.382 2.382
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6178 0.0932 28.08 <0.0000000000000002 ***
## challenges_trustYes 0.4330 0.1784 2.43 0.016 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.17 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.0268, Adjusted R-squared: 0.0223
## F-statistic: 5.89 on 1 and 214 DF, p-value: 0.016Q12 Pred: None
Business pressures (e.g., speed to market)
##
## Call:
## lm(formula = Q13_1 ~ challenges_none, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.831 -0.831 0.169 0.831 1.831
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.8311 0.0726 52.76 < 0.0000000000000002 ***
## challenges_noneYes -0.6624 0.1385 -4.78 0.0000027 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.07 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0708, Adjusted R-squared: 0.0677
## F-statistic: 22.9 on 1 and 300 DF, p-value: 0.00000272Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_2 ~ challenges_none, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.657 -0.657 -0.422 0.578 2.578
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.658 0.074 35.92 <0.0000000000000002 ***
## challenges_noneYes -0.236 0.141 -1.67 0.096 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.09 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00923, Adjusted R-squared: 0.00592
## F-statistic: 2.79 on 1 and 300 DF, p-value: 0.0957Concerns about being labeled a “troublemaker”
##
## Call:
## lm(formula = Q13_3 ~ challenges_none, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.0872 -1.0872 -0.0872 0.9128 2.5301
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.0872 0.0753 41.0 < 0.0000000000000002 ***
## challenges_noneYes -0.6173 0.1435 -4.3 0.000023 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.11 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0583, Adjusted R-squared: 0.0551
## F-statistic: 18.5 on 1 and 299 DF, p-value: 0.0000229Other
##
## Call:
## lm(formula = Q13_4 ~ challenges_none, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.829 -0.829 0.171 0.484 2.484
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8289 0.0953 29.68 <0.0000000000000002 ***
## challenges_noneYes -0.3133 0.1751 -1.79 0.075 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.18 on 214 degrees of freedom
## (87 observations deleted due to missingness)
## Multiple R-squared: 0.0147, Adjusted R-squared: 0.0101
## F-statistic: 3.2 on 1 and 214 DF, p-value: 0.075Q13 Business pressures (e.g., speed to market)
Take ethical / responsible AI trainings
##
## Call:
## lm(formula = q9_takeethical ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.454 -0.424 -0.414 0.576 0.586
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.46358 0.09802 4.73 0.0000035 ***
## Q13_1 -0.00998 0.02570 -0.39 0.7
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.496 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000503, Adjusted R-squared: -0.00283
## F-statistic: 0.151 on 1 and 300 DF, p-value: 0.698Conduct audits of the gen AI tools
##
## Call:
## lm(formula = q9_conductaudits ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.295 -0.265 -0.236 0.705 0.823
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1470 0.0861 1.71 0.089 .
## Q13_1 0.0296 0.0226 1.31 0.191
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.436 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00569, Adjusted R-squared: 0.00237
## F-statistic: 1.72 on 1 and 300 DF, p-value: 0.191Conduct fairness / bias testing
##
## Call:
## lm(formula = q9_conductfair ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.323 -0.307 -0.291 0.677 0.741
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2432 0.0909 2.68 0.0079 **
## Q13_1 0.0159 0.0238 0.67 0.5041
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.46 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00149, Adjusted R-squared: -0.00184
## F-statistic: 0.447 on 1 and 300 DF, p-value: 0.504Consider data privacy implications and take actions to protect data privacy
##
## Call:
## lm(formula = q9_considerdatapriv ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.495 -0.477 -0.431 0.523 0.578
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4032 0.0988 4.08 0.000058 ***
## Q13_1 0.0184 0.0259 0.71 0.48
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00167, Adjusted R-squared: -0.00166
## F-statistic: 0.501 on 1 and 300 DF, p-value: 0.479Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## lm(formula = q9_askaboutdata ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.448 -0.401 -0.353 0.599 0.742
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2107 0.0958 2.20 0.029 *
## Q13_1 0.0475 0.0251 1.89 0.060 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.485 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0118, Adjusted R-squared: 0.00849
## F-statistic: 3.58 on 1 and 300 DF, p-value: 0.0595Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## lm(formula = q9_adversarial ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.1971 -0.1566 -0.1566 -0.0756 0.9649
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.00535 0.06869 -0.08 0.938
## Q13_1 0.04049 0.01801 2.25 0.025 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.348 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.0166, Adjusted R-squared: 0.0133
## F-statistic: 5.05 on 1 and 300 DF, p-value: 0.0253Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## lm(formula = q9_explainability ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.342 -0.314 -0.287 0.658 0.768
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2039 0.0910 2.24 0.026 *
## Q13_1 0.0276 0.0239 1.16 0.248
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.461 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00444, Adjusted R-squared: 0.00112
## F-statistic: 1.34 on 1 and 300 DF, p-value: 0.248Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## lm(formula = q9_buildtransparency ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.361 -0.322 -0.282 0.639 0.796
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1646 0.0911 1.81 0.072 .
## Q13_1 0.0393 0.0239 1.64 0.101
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.461 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00894, Adjusted R-squared: 0.00563
## F-statistic: 2.7 on 1 and 300 DF, p-value: 0.101No actions taken (to my knowledge)
##
## Call:
## lm(formula = q9_noactions ~ Q13_1, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.319 -0.183 -0.183 -0.137 0.863
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3645 0.0785 4.65 0.0000051 ***
## Q13_1 -0.0454 0.0206 -2.21 0.028 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.397 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.016, Adjusted R-squared: 0.0127
## F-statistic: 4.88 on 1 and 300 DF, p-value: 0.0279Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.601 0.443 1.36 0.18
## Q13_1 0.166 0.119 1.40 0.16
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 327.02 on 301 degrees of freedom
## Residual deviance: 325.10 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 329.1
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.73## 2.5 % 97.5 %
## 0.9324 1.4895Lack of training
##
## Call:
## glm(formula = challenges_training ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.7113 0.4104 -1.73 0.083 .
## Q13_1 0.0768 0.1071 0.72 0.474
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 404.99 on 301 degrees of freedom
## Residual deviance: 404.48 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 408.5
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.08## 2.5 % 97.5 %
## 0.8769 1.3364Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.641 0.401 1.60 0.110
## Q13_1 -0.216 0.105 -2.05 0.041 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 417.06 on 301 degrees of freedom
## Residual deviance: 412.79 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 416.8
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 0.8059## 2.5 % 97.5 %
## 0.6534 0.9891Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.649 0.516 -5.13 0.00000028 ***
## Q13_1 0.565 0.130 4.35 0.00001340 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 398.29 on 301 degrees of freedom
## Residual deviance: 375.91 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 379.9
##
## Number of Fisher Scoring iterations: 3## Q13_1
## 1.759## 2.5 % 97.5 %
## 1.377 2.295Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.727 0.590 -4.63 0.0000037 ***
## Q13_1 0.374 0.147 2.55 0.011 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 311.96 on 301 degrees of freedom
## Residual deviance: 304.68 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 308.7
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.454## 2.5 % 97.5 %
## 1.377 2.295Lack of support
##
## Call:
## glm(formula = challenges_support ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.337 0.582 -4.01 0.00006 ***
## Q13_1 0.212 0.147 1.44 0.15
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.56 on 301 degrees of freedom
## Residual deviance: 278.37 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 282.4
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.236## 2.5 % 97.5 %
## 0.9355 1.6718Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.101 0.564 -3.72 0.0002 ***
## Q13_1 0.143 0.144 0.99 0.3211
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.44 on 301 degrees of freedom
## Residual deviance: 276.42 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 280.4
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.154## 2.5 % 97.5 %
## 0.8777 1.5502Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ Q13_1, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -3.089 0.588 -5.25 0.00000015 ***
## Q13_1 0.548 0.145 3.78 0.00016 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.21 on 301 degrees of freedom
## Residual deviance: 334.21 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 338.2
##
## Number of Fisher Scoring iterations: 4## Q13_1
## 1.73## 2.5 % 97.5 %
## 1.318 2.332Q13 Concerns about being labeled a “troublemaker”
Take ethical / responsible AI trainings
##
## Call:
## lm(formula = q9_takeethical ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.470 -0.434 -0.399 0.566 0.602
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3804 0.0733 5.19 0.00000039 ***
## Q13_2 0.0180 0.0260 0.69 0.49
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.496 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00159, Adjusted R-squared: -0.00173
## F-statistic: 0.479 on 1 and 300 DF, p-value: 0.489Conduct audits of the gen AI tools
##
## Call:
## lm(formula = q9_conductaudits ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.297 -0.271 -0.244 0.703 0.809
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3238 0.0645 5.02 0.00000088 ***
## Q13_2 -0.0266 0.0229 -1.16 0.25
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.436 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00446, Adjusted R-squared: 0.00114
## F-statistic: 1.34 on 1 and 300 DF, p-value: 0.247Conduct fairness / bias testing
##
## Call:
## lm(formula = q9_conductfair ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.308 -0.302 -0.300 0.695 0.703
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.29373 0.06802 4.32 0.000021 ***
## Q13_2 0.00293 0.02417 0.12 0.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.46 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 4.9e-05, Adjusted R-squared: -0.00328
## F-statistic: 0.0147 on 1 and 300 DF, p-value: 0.904Consider data privacy implications and take actions to protect data privacy
##
## Call:
## lm(formula = q9_considerdatapriv ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.519 -0.488 -0.411 0.512 0.604
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.5500 0.0738 7.45 0.000000000001 ***
## Q13_2 -0.0308 0.0262 -1.17 0.24
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00457, Adjusted R-squared: 0.00125
## F-statistic: 1.38 on 1 and 300 DF, p-value: 0.242Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## lm(formula = q9_askaboutdata ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.414 -0.395 -0.358 0.605 0.661
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4324 0.0720 6.00 0.0000000056 ***
## Q13_2 -0.0186 0.0256 -0.73 0.47
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.488 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00176, Adjusted R-squared: -0.00157
## F-statistic: 0.529 on 1 and 300 DF, p-value: 0.468Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## lm(formula = q9_adversarial ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.149 -0.145 -0.141 -0.137 0.868
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1530 0.0518 2.95 0.0034 **
## Q13_2 -0.0041 0.0184 -0.22 0.8240
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.351 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000165, Adjusted R-squared: -0.00317
## F-statistic: 0.0496 on 1 and 300 DF, p-value: 0.824Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## lm(formula = q9_explainability ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.325 -0.312 -0.300 0.685 0.725
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3370 0.0682 4.94 0.0000013 ***
## Q13_2 -0.0125 0.0242 -0.52 0.61
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.462 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.000884, Adjusted R-squared: -0.00245
## F-statistic: 0.265 on 1 and 300 DF, p-value: 0.607Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## lm(formula = q9_buildtransparency ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.314 -0.309 -0.306 0.689 0.696
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.30168 0.06844 4.41 0.000015 ***
## Q13_2 0.00242 0.02431 0.10 0.92
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.463 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 3.3e-05, Adjusted R-squared: -0.0033
## F-statistic: 0.00989 on 1 and 300 DF, p-value: 0.921No actions taken (to my knowledge)
##
## Call:
## lm(formula = q9_noactions ~ Q13_2, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.228 -0.204 -0.191 -0.179 0.821
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1670 0.0591 2.82 0.0051 **
## Q13_2 0.0122 0.0210 0.58 0.5609
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00113, Adjusted R-squared: -0.0022
## F-statistic: 0.339 on 1 and 300 DF, p-value: 0.561Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.4600 0.3548 4.11 0.000039 ***
## Q13_2 -0.0997 0.1233 -0.81 0.42
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 327.02 on 301 degrees of freedom
## Residual deviance: 326.37 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 330.4
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 0.9999## 2.5 % 97.5 %
## 0.711 1.155Lack of training
##
## Call:
## glm(formula = challenges_training ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.927 0.308 -3.01 0.0026 **
## Q13_2 0.190 0.108 1.76 0.0782 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 404.99 on 301 degrees of freedom
## Residual deviance: 401.87 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 405.9
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 1.209## 2.5 % 97.5 %
## 0.9795 1.4968Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.0316 0.2963 -0.11 0.92
## Q13_2 -0.0442 0.1054 -0.42 0.68
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 417.06 on 301 degrees of freedom
## Residual deviance: 416.88 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 420.9
##
## Number of Fisher Scoring iterations: 3## Q13_2
## 0.9568## 2.5 % 97.5 %
## 0.7772 1.1762Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.886 0.310 -2.86 0.0042 **
## Q13_2 0.137 0.109 1.26 0.2079
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 398.29 on 301 degrees of freedom
## Residual deviance: 396.70 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 400.7
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 1.146## 2.5 % 97.5 %
## 0.9268 1.4200Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.790 0.372 -4.81 0.0000015 ***
## Q13_2 0.180 0.127 1.42 0.16
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 311.96 on 301 degrees of freedom
## Residual deviance: 309.96 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 314
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 1.197## 2.5 % 97.5 %
## 0.9268 1.4200Lack of support
##
## Call:
## glm(formula = challenges_support ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.227 0.408 -5.46 0.000000047 ***
## Q13_2 0.253 0.136 1.86 0.063 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 280.56 on 301 degrees of freedom
## Residual deviance: 277.10 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 281.1
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 1.287## 2.5 % 97.5 %
## 0.9861 1.6815Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.291 0.412 -5.56 0.000000028 ***
## Q13_2 0.267 0.137 1.96 0.05 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.44 on 301 degrees of freedom
## Residual deviance: 273.63 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 277.6
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 1.306## 2.5 % 97.5 %
## 0.9989 1.7102Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ Q13_2, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.003352 0.333530 -3.01 0.0026 **
## Q13_2 -0.000139 0.118489 0.00 0.9991
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 351.21 on 301 degrees of freedom
## Residual deviance: 351.21 on 300 degrees of freedom
## (1 observation deleted due to missingness)
## AIC: 355.2
##
## Number of Fisher Scoring iterations: 4## Q13_2
## 0.9999## 2.5 % 97.5 %
## 0.7906 1.2597Q13 Discomfort raising issues around responsibility
Take ethical / responsible AI trainings
##
## Call:
## lm(formula = q9_takeethical ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.460 -0.427 -0.410 0.556 0.607
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3760 0.0783 4.80 0.0000025 ***
## Q13_3 0.0169 0.0250 0.67 0.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.496 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00152, Adjusted R-squared: -0.00182
## F-statistic: 0.456 on 1 and 299 DF, p-value: 0.5Conduct audits of the gen AI tools
##
## Call:
## lm(formula = q9_conductaudits ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.332 -0.292 -0.213 0.668 0.827
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3717 0.0688 5.40 0.00000014 ***
## Q13_3 -0.0397 0.0220 -1.81 0.072 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.435 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.0108, Adjusted R-squared: 0.00751
## F-statistic: 3.27 on 1 and 299 DF, p-value: 0.0716Conduct fairness / bias testing
##
## Call:
## lm(formula = q9_conductfair ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.353 -0.304 -0.280 0.671 0.744
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.2314 0.0727 3.18 0.0016 **
## Q13_3 0.0243 0.0232 1.05 0.2953
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.46 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00366, Adjusted R-squared: 0.00033
## F-statistic: 1.1 on 1 and 299 DF, p-value: 0.295Consider data privacy implications and take actions to protect data privacy
##
## Call:
## lm(formula = q9_considerdatapriv ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.548 -0.465 -0.382 0.535 0.618
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.5894 0.0788 7.48 0.00000000000082 ***
## Q13_3 -0.0415 0.0251 -1.65 0.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.498 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00901, Adjusted R-squared: 0.0057
## F-statistic: 2.72 on 1 and 299 DF, p-value: 0.1Ask about the data or model to understand potential limitations or issues it has
##
## Call:
## lm(formula = q9_askaboutdata ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.418 -0.381 -0.361 0.601 0.657
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4374 0.0770 5.68 0.000000032 ***
## Q13_3 -0.0190 0.0246 -0.77 0.44
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.487 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00199, Adjusted R-squared: -0.00135
## F-statistic: 0.595 on 1 and 299 DF, p-value: 0.441Conduct adversarial testing or red teaming (pretend to be a bad actor to test how a product we develop could be used for harm by bad actors)
##
## Call:
## lm(formula = q9_adversarial ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.147 -0.143 -0.139 -0.135 0.869
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.15076 0.05494 2.74 0.0064 **
## Q13_3 -0.00385 0.01754 -0.22 0.8265
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.348 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.000161, Adjusted R-squared: -0.00318
## F-statistic: 0.0481 on 1 and 299 DF, p-value: 0.827Use explainability methods (e.g., to enable team members to better understand and probe the model)
##
## Call:
## lm(formula = q9_explainability ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.321 -0.303 -0.294 0.688 0.715
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.27591 0.07281 3.79 0.00018 ***
## Q13_3 0.00906 0.02324 0.39 0.69708
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.461 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.000508, Adjusted R-squared: -0.00284
## F-statistic: 0.152 on 1 and 299 DF, p-value: 0.697Build transparency approaches (e.g., use documentation that can make dataset and model decisions transparent to others)
##
## Call:
## lm(formula = q9_buildtransparency ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.314 -0.306 -0.302 0.690 0.702
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.29346 0.07305 4.02 0.000074 ***
## Q13_3 0.00418 0.02332 0.18 0.86
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.462 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.000107, Adjusted R-squared: -0.00324
## F-statistic: 0.0321 on 1 and 299 DF, p-value: 0.858No actions taken (to my knowledge)
##
## Call:
## lm(formula = q9_noactions ~ Q13_3, data = genaiclean)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.252 -0.227 -0.176 -0.151 0.849
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.1252 0.0632 1.98 0.048 *
## Q13_3 0.0254 0.0202 1.26 0.209
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.00528, Adjusted R-squared: 0.00195
## F-statistic: 1.59 on 1 and 299 DF, p-value: 0.209Lack of clarity on what that looks like
##
## Call:
## glm(formula = challenges_clarity ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.3538 0.3789 3.57 0.00035 ***
## Q13_3 -0.0545 0.1196 -0.46 0.64890
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 326.49 on 300 degrees of freedom
## Residual deviance: 326.29 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 330.3
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.297## 2.5 % 97.5 %
## 0.7478 1.1969Lack of training
##
## Call:
## glm(formula = challenges_training ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.371 0.343 -4.00 0.000064 ***
## Q13_3 0.315 0.107 2.94 0.0032 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 403.13 on 300 degrees of freedom
## Residual deviance: 394.16 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 398.2
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.37## 2.5 % 97.5 %
## 1.114 1.695Lack of resources or tools
##
## Call:
## glm(formula = challenges_resources ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.523 0.320 1.63 0.102
## Q13_3 -0.233 0.103 -2.26 0.024 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 415.52 on 300 degrees of freedom
## Residual deviance: 410.32 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 414.3
##
## Number of Fisher Scoring iterations: 4## <NA>
## NA## 2.5 % 97.5 %
## 0.6461 0.9682Lack of incentives
##
## Call:
## glm(formula = challenges_incentives ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.321 0.344 -3.84 0.00012 ***
## Q13_3 0.269 0.107 2.52 0.01186 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 397.36 on 300 degrees of freedom
## Residual deviance: 390.86 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 394.9
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.309## 2.5 % 97.5 %
## 1.064 1.620Lack of understanding whether / why it may be valuable
##
## Call:
## glm(formula = challenges_valuable ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.923 0.410 -4.69 0.0000028 ***
## Q13_3 0.205 0.125 1.63 0.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 311.48 on 300 degrees of freedom
## Residual deviance: 308.77 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 312.8
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.227## 2.5 % 97.5 %
## 1.064 1.620Lack of support
##
## Call:
## glm(formula = challenges_support ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.850 0.488 -5.84 0.0000000052 ***
## Q13_3 0.415 0.142 2.92 0.0034 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.06 on 300 degrees of freedom
## Residual deviance: 267.98 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 272
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.515## 2.5 % 97.5 %
## 1.154 2.017Lack of clarity about expectations or relevance to my role
##
## Call:
## glm(formula = challenges_clarityexpectations ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.472 0.463 -5.34 0.000000095 ***
## Q13_3 0.298 0.138 2.16 0.031 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 277.06 on 300 degrees of freedom
## Residual deviance: 272.25 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 276.3
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.347## 2.5 % 97.5 %
## 1.032 1.776Lack of trust (personally)in the gen AI tool/model itself
##
## Call:
## glm(formula = challenges_trust ~ Q13_3, family = "binomial",
## data = genaiclean)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.778 0.382 -4.66 0.0000032 ***
## Q13_3 0.260 0.117 2.23 0.026 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 350.58 on 300 degrees of freedom
## Residual deviance: 345.50 on 299 degrees of freedom
## (2 observations deleted due to missingness)
## AIC: 349.5
##
## Number of Fisher Scoring iterations: 4## Q13_3
## 1.297## 2.5 % 97.5 %
## 1.034 1.636