Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training
Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated.
More than 1,000 students were randomized to test ChatGPT and critical-thinking training on a real university assignment, with outcomes including performance, originality, and critical thinking. For teams shipping education-focused LLM tools, the practical takeaway is that access alone is not the product: measurable learning gains need to be evaluated in authentic workflows and paired with explicit thinking scaffolds.
Using ChatGPT with critical-thinking training improved student performance on university assignments, demonstrating that LLMs can enhance outcomes when paired with structured guidance. This highlights the importance of integrating training frameworks alongside LLMs in production to maximize effectiveness and mitigate risks of over-reliance or shallow outputs.