When AI Meets RI: Legal and Compliance Considerations Emerging from the Use of AI Tools to Identify Research Integrity Concerns This is the fifth in a series of posts addressing the use of AI-based screening tools to detect research integrity issues, including pros and cons, unique challenges, downstream implications, potential pitfalls, and risk mitigation practices. Through our experience with research integrity matters, we have identified a series of risk mitigation practices that institutions may wish to consider: Broad Access to AI Tools (with Guardrails) Provide research teams with quality AI image and plagiarism detection tools, along with clear instructions that they are not substitutes for careful author review. Consider mandating use for manuscripts and grants prior to formal submission. Regular Training Conduct thorough (in-person, when feasible) initial and periodic refresher training on research misconduct definitions, common drivers of misconduct, and prevention practices. Strengthen PI Oversight Foster and evaluate mentoring, supervision, and leadership. Ensure that PIs and supervisors (1) are trained, equipped, and executing on their responsibilities, and (2) establish clear expectations for their laboratories. Tighten Data Storage and Retention Efforts Establish explicit standards for research data storage, retention, transfer, and access. Consider requiring use of electronic laboratory notebooks. Conduct periodic audits. Correct the Record (Timely AND Accurately) Encourage timely corrections and retractions when warranted. Accuracy is paramount – haste in preparing and submitting a correction can make things worse. Create Robust Onboarding Processes Conduct in-person onboarding, supplemented with written guidance, to improve compliance with expectations, policies, and procedures. Formal/Centralized Offboarding Establish policies and processes for project handoffs, access termination, equipment management, and materials/data transfer.