Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension

AI Generated Summary*

CONSORT-AI, produced in parallel with the protocol-focused SPIRIT-AI, adds items to the CONSORT 2010 checklist for reports of randomized trials testing interventions that include an artificial intelligence component, addressing bias sources specific to AI that earlier guidance left uncovered. Registry and literature review plus expert consultation produced 29 candidate items, rated across two Delphi rounds (103 first-round respondents), voted on by 31 stakeholders over two days against a pre-set 80% inclusion threshold, then piloted with 34 people. Fourteen items resulted (11 extensions, 3 elaborations), spanning intended use, algorithm version, input data eligibility and quality handling, site integration, human-AI interaction, outputs and their role in decisions, performance error analysis, and code access. Continuously evolving systems were left out for now.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.