Smarter or Just Faster?

University students now use AI more than ever. But are they learning more — or just offloading the hard parts?

In just three years, an entire generation has adopted a new study partner — one that never sleeps and can write an essay in under a minute. The question is no longer whether students use AI. It is what it is doing to them.

“When thinking becomes optional, does learning still happen?”

The surge

AI adoption among university students, 2022–2025 — hover for details

Source: Freeman (2024, 2025) Student generative AI surveys (HEPI Policy Notes 51 & 61). Higher Education Policy Institute.

By 2025, nine in ten students use AI, up from one in three just three years ago (Freeman, 2025). But adoption tells us nothing about impact. What matters is how students are using it.

What students are actually doing

Task breakdown by use type, 2024 vs 2025 — hover to compare

Source: (Freeman,2025) Student generative AI survey 2025 (HEPI Policy Note 61). Higher Education Policy Institute.

Explaining concepts is the most common task at 58%, yet 18% now submit AI-generated text directly in assessments (Freeman, 2025). AI amplifies existing tendencies rather than reshaping them. So why do students turn to it and what holds them back?

“AI is a mirror. It reflects back whatever approach you already bring to studying.”

The tension: why they use it — and what holds them back

Motivations vs barriers, 2025 — hover each bar

Source: (Freeman, 2025) Student generative AI survey 2025 (HEPI Policy Note 61). Higher Education Policy Institute.

The motivations are deeply human — 51% say it saves time, 50% say it improves their work. Yet more students fear being accused of cheating (53%) than are motivated by any single benefit (Freeman, 2025). And not every student faces this tension equally.

The AI divide

AI use by field, gender and socioeconomic background — use buttons to filter, hover each point

Source: (Freeman, 2025) Student generative AI survey 2025 (HEPI Policy Note 61). Higher Education Policy Institute.

Male students are 14 percentage points more likely to use AI than female students, and higher socioeconomic groups show greater adoption across every field (Freeman, 2025). If AI benefits most those already most advantaged, it risks widening the inequalities it could have closed.

“If AI is a superpower, who already has access to the cape?”

Growing use, growing doubt

AI homework use vs student concern about critical thinking, 2025 — hover to explore

Source: (Schwartz & Diliberti, 2026) More students use AI for homework and more believe it harms critical thinking.

As AI use climbed from 48% to 62% across 2025, so did student concern — by December, 67% believed AI was harming their peers’ critical thinking, up from 54% ten months earlier (Schwartz & Diliberti, 2026). Students are using a tool they are increasingly uncertain about, with little institutional guidance on how to use it well.

The debate has focused on whether to allow AI. The data points to something more urgent: how do we teach students to use it in ways that make them more capable — not just faster — human beings?

References

Freeman, J. (2024). Provide or punish? Students’ views on generative AI in higher education. https://www.hepi.ac.uk/wp-content/uploads/2024/01/HEPI- Policy-Note-51.pdf

Freeman, J. (2025). Student Generative AI Survey 2025. https://www.hepi.ac.uk/wp-content/uploads/2025/02/HEPI- Kortext-Student-Generative-AI-Survey-2025.pdf

Schwartz, H. L., & Diliberti, M. K. (2026). More Students Use AI for Homework, and More Believe It Harms Critical Thinking: Selected Findings from the American Youth Panel. Rand.org; RAND Corporation. https://www.rand.org/pubs/research_reports/RRA4742-1.html

Generative AI acknowledgement: No AI tools were used in the preparation of this assignment. All data was sourced independently from the references listed above. All visualisations were created using R.