Day 5 · Week 1 · AI & self-care
There is a source link. I would open it.
A row of references looks reassuring. For my own health, I would choose one claim that could affect a decision and actually check it.
I would enter that claim in the prompt below and ask AI to open a reliable source and identify the supporting section. If it could not open it, I would want it to say so. I would open the link myself, checking that the page existed, addressed the same claim and described the relevant population.
A prompt I'd try:
Copy the prompt, then open ChatGPT and paste it into a new chat.
Check this medical claim: [claim]. Context: [general topic, if useful]. If you can browse, open a reliable medical source and give its title, link, publication or update date if verified, and the section supporting or contradicting the claim. Explain what the source does not establish. If you can't open or verify it, say so; don't invent a citation or pretend you checked. Keep it under 150 words. One source is a starting point, not proof that the whole answer is correct.
Copy manually
Press and hold the text below, choose Copy, then paste it in ChatGPT.
For example, “prevents cancer” and “lowers a blood-test number” are different claims. I would compare the actual wording and save any unresolved question and link for my doctor or pharmacist.
I would not need to read every paper overnight. Checking one decision-relevant claim would beat collecting ten unopened links, without proving the rest of the answer correct.