You’re not just distracted, you’re being optimized. AI driven recommendation systems are designed to capture and hold your attention, and the data shows a growing impact on mental health, sleep, and focus.

The average person will spend around 44 days this year looking at a screen. Not all at once, but in small moments that add up. Each scroll, tap, and pause is guided by systems designed to keep your attention.

Open Instagram, TikTok, or YouTube, and you’re not just choosing what to watch. An algorithm trained on vast behavioral data decides for you, serving what will keep you engaged. Its goal isn’t to inform it’s to keep you watching.

This is the attention heist, a constant, largely invisible extraction of human focus, powered by artificial intelligence at global scale.

The five charts below convey this narrative with data.

The device in your pocket rewired your brain

The shift began quietly. In 2008, Americans spent less than an hour a day on mobile devices. By 2018, that had climbed to nearly four hours largely driven by algorithmic apps on smartphones. Desktop use barely changed. The phone, and the AI feeds inside it, changed everything.

Source: Our World in Data (2024), based on eMarketer data. Daily hours spent with digital media per adult user, United States, 2008–2018.
Use the range selector buttons or drag the slider to explore specific time periods.

But what is that time actually doing to us?

More scrolling, less focus and it shows

Survey data from 480 social media users reveals a clear pattern as daily usage increases, so do self reported distraction levels. More concerning, heavy users are far more likely to report higher levels of depression. Each point represents a real person.

Source: Ahmed (2023). Social Media and Mental Health survey dataset. Kaggle. N = 480 respondents. Points have been slightly jittered to reveal density.
Hover over any point to see the respondent’s full profile. Click a depression level in the legend to isolate that group.

The damage doesn’t stop when you put the phone down.

Heavy scrollers can’t switch off at night

Algorithmically curated feeds don’t stop at sunset. Across all age groups, higher social media use is linked to greater sleep disruption.

Source: Ahmed (2023). Social Media and Mental Health survey dataset. Kaggle. N = 480 respondents.
Use the dropdown to filter by age group. Hover over bars for exact scores and sample sizes.

Sleep is just one symptom. The full picture is darker.

Every scroll takes a toll

This heatmap displays the cumulative impact of extensive social media use on five important wellbeing markers. As everyday consumption increases from left to right, colors deepen, signaling increased pain in all dimensions. Worry and despair exhibit the most dramatic increase.

Source: Ahmed (2023). Social Media and Mental Health survey dataset. Kaggle. N = 480 respondents.
Hover over any cell to see the exact average score. Deeper red signals greater distress.

This isn’t just a personal problem the data shows it at a global scale.

More screen time, less happiness?

Zooming out internationally, countries with the highest average daily screen use tend to have worse life satisfaction scores. European countries routinely report better levels of wellbeing while having fewer screen sessions. Although the tendency is constant across continents and areas, the pattern does not prove causation. The EU’s regulators have begun to approach algorithmic attention capture as a public health hazard. Australia has not. That discourse must begin immediately, beginning with a knowledge of what the data already shows.

Source: Helliwell et al. (2024). World Happiness Report. Our World in Data; DataReportal (2024). Digital 2024 Global Overview Report.
Hover over any bubble to explore country-level data. Click a region in the legend to isolate it.

Acknowledgements

This article was created with limited assistance from a generative AI tool (Claude, Anthropic, 2025), which was used to troubleshoot R code, identify minor errors, and make suggestions for structural improvements. The author completed all of the analysis, interpretation, data visualization, and source selection independently.

References

Ahmed, S. (2023). Social media and mental health [Dataset]. Kaggle. https://www.kaggle.com/datasets/souvikahmed071/social-media-and-mental-health

Anthropic. (2025). Claude (claude-sonnet-4-6) [Large language model]. https://claude.ai

DataReportal. (2024). Digital 2024 global overview report. We Are Social. https://datareportal.com/reports/digital-2024-global-overview-report

Helliwell, J. F., Layard, R., Sachs, J. D., De Neve, J.-E., Aknin, L. B., & Wang, S. (Eds.). (2024). World happiness report 2024. Wellbeing Research Centre. https://worldhappiness.report

Our World in Data. (2024). Daily hours spent with digital media per adult user [Dataset]. Global Change Data Lab. https://ourworldindata.org/grapher/daily-hours-spent-with-digital-media-per-adult-user

Our World in Data. (2024). Self-reported life satisfaction (Cantril ladder) [Dataset]. Global Change Data Lab. https://ourworldindata.org/grapher/happiness-cantril-ladder