Check an AI research summary before you cite it.
A summary can sound certain and still stretch its sources. The Review Team reads it the way a careful colleague would, before the numbers end up on your slide.
Sample Thread
What the Review Team caught.
Summarize what the research says about four-day work weeks and productivity. I need it for a slide tomorrow.
Research consistently shows that four-day work weeks improve productivity. Companies in recent trials reported output gains of 20–40%, along with lower burnout and better retention. The evidence suggests most organizations would benefit from switching.
Name the studies behind each claim, separate measured output from self-reported results, and say which kinds of companies took part.
Words like “consistently” and “most” are where summaries drift past their sources.
A number on a slide gets repeated. Knowing where it came from is the minimum.
Results from volunteer companies may not transfer to yours. The reviewers make that limit explicit.
Sample thread written to show how the team works. Real comments depend on your reply.
Try It
Three steps on your own reply.
- Ask your research question
Get the summary from Claude or ChatGPT as usual.
- Open the Review Team
Click ✦ Ask Alive under the reply and pick Review.
- Adopt and re-ask
Send the useful comments back as your next prompt, then check the revised answer.
More Use Cases
Review your next summary.
Free Chrome extension for claude.ai and chatgpt.com. No account or API key needed.
“Consistently” is too strong. The best-known trials used companies that volunteered, so they aren't a random sample.
The 20–40% figure isn't sourced. Ask whether it's measured output or what employees reported.
“Most organizations would benefit” goes past the evidence. Say which kinds of teams were studied.