Global evidence leaders mandate responsible AI use

In a unified response to the rapid rise of Artificial Intelligence (AI) and its use, global organisations in evidence synthesis, Cochrane, the Campbell Collaboration, JBI, and the Collaboration for Environmental Evidence (CEE), have co-published a position statement defining clear standards for the responsible use of AI in evidence syntheses.

The position statement was simultaneously released today across four journals: JBI Evidence Synthesis, Campbell Systematic Reviews, Cochrane Database of Systematic Reviews, and Environmental Evidence, marking a unified global approach to managing AI's integration into evidence-based research.

The position statement and its co-publication stems is the result of work undertaken by the Joint AI Methods Group, a collaboration between Cochrane, Campbell Collaboration, JBI and CEE.

The Joint AI Methods Group officially supports the Responsible use of AI in evidence SynthEsis (RAISE) recommendations, which provide a framework for ensuring responsible AI use across the evidence synthesis ecosystem.

The authors, all members of the Joint AI Methods Group, acknowledge that AI, ranging from rule-based algorithms to large language models and generative AI, offers significant potential to transform evidence synthesis by making the process more efficient, timely, and affordable. However, they warn that the technology poses serious risks to research integrity.

"AI is characterized by opaque decision-making, susceptible to algorithmic biases, and carries the risk of fabricated outputs and hallucinations," the statement cautions. These factors could compromise the methodological rigour that evidence syntheses are built upon—transparency, reproducibility, and reliability.

The concern is heightened by the fact that many AI developments are commercially driven, often lacking appropriate validation and transparency regarding their limitations.

The position statement outlines six mandates, beginning with evidence synthesists remaining accountable for their work, including all decisions about AI use and adherence to ethical and legal standards.

Zoe Jordan, Executive Director of JBI, is one of the position statement's authors and a member of the Joint AI Methods Group. Prof Jordan emphasised the balance between innovation and rigour:

"Evidence synthesists must be able to justify the specific use of any AI tool in their context, demonstrating that it is methodologically sound and will not compromise the conclusions of the synthesis."

The Joint AI Methods Group stresses that transparent information from AI developers is essential for evidence synthesists to make informed decisions. This includes understanding the scope and domains of training data, potential biases such as English-only or open-access-only datasets, and comprehensive performance evaluations.

The position statement, Position statement on artificial intelligence (AI) use in evidence synthesis, is available in the November 2025 issue of JBI Evidence Synthesis.