One review model. Many domains.

Review intended use, training data, evaluation, bias, robustness, limitations, human oversight, user disclosure, and monitoring. For autonomous agents, review tool permissions, delegated authority, credential scope, revocation, execution limits, and chain-of-custody.

Healthcare AI demands a higher standard because the consequences touch safety, trust, dignity, access, and life itself. OpenIRB supports structured review of clinical AI, patient data, medical workflows, and research-adjacent systems.

Review participant protections, consent, privacy, data reuse, AI-assisted analysis, risk-benefit balance, and institutional accountability.

Review systems affecting benefits, enforcement, identity, public safety, services, transparency, appeal rights, and human oversight.

Review student data, assessment, personalization, developmental impact, fairness, accessibility, and child protections.

Review AI projects across lifecycle gates, SOPs, leadership accountability, deployment readiness, and monitoring.