Comparison of AI-assisted and human-generated plain language summaries for Cochrane reviews: a randomised non-inferiority trial (HIET-1)
Evidence Synthesis Ireland contributed to the development of the registered report of Comparison of AI-assisted and human-generated plain language summaries for Cochrane reviews: a randomised non-inferiority trial (HIET-1). This report compares the comprehension, readability, quality, safety, and trustworthiness of artificial intelligence (AI)-assisted vs human-generated plain language summaries (PLSs) for Cochrane systematic reviews.
Authors:
Declan Devane, Johanna Pope, Paula Byrne, Evan Forde, Isabel O’Byrne, Steven Woloshin, Eileen Culloty, Darren Dahly, Ingeborg Hess Elgersma, Heather Munthe-Kaas, Conor Judge, Martin O’Donnell, Finn Krewer, Sandra Galvin, Nikita N. Burke, Theresa Tierney, KM Saif-Ur-Rahman, Tom Conway, James Thomas.
Key findings:
• AI summaries achieved noninferior comprehension vs human-generated versions.
• Human verification corrected critical numerical errors in AI drafts.
• Randomised trial testing public comprehension of AI health summaries.
Read the full report here.
