When students ask an AI tool to summarise a reading, it is tempting to accept the first answer. That answer can sound convincing while being inaccurate, oversimplified or incomplete, and many students have not yet built the habit of checking it against the original source.

Fanny Tsai asks students to summarise the same scholarly source with three different generative AI tools. They compare the outputs for accuracy, clarity, completeness and usefulness, and then verify the most important claims against the original text.

Using their own disciplinary knowledge, students then create a short explainer of one key concept, claim or question for an intended audience, as a written response, narrated slides or a video. This step asks them to put the ideas into their own words instead of passing on an AI summary.

Finally, students review the explainers of two classmates with the CAAA criteria: clarity, accuracy, accessibility and audience alignment. Each review names one strength and one specific suggestion, and instructors can adapt the source, tools, audience and format to any discipline or level.

Students learn to use AI as a starting point rather than a final answer, while practising source verification, clear communication and constructive peer feedback in one activity.