Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
MIT AI Incident Trackerairisk.mit.edu · Low evidenceThe MIT AI Incident Tracker classifies over 1,400 reported AI incidents by risk, cause, harm, and severity using a Large Language Model. It applies established frameworks like the MIT Risk Repository and CSET's AI Harm Taxonomy to help researchers, policymakers, and regulators understand AI risks.
MIT AI Incident TrackerParse Score
