MIT Launches ImpactBench, First Open Benchmark Measuring AI's Impact on Human Well-Being
LuizaJarovsky · x · 2026-10-08
MIT Media Lab and partner research teams have released ImpactBench, billed as the first open benchmark measuring AI's impact on human well-being across physical, psychological, and social dimensions.
The benchmark examines core questions: whether AI helps people learn or does the thinking for them, supports mental health or encourages harm, strengthens or replaces creativity, respects or overrides human decisions, and fosters human connection or dependence on itself — each with multiple metrics.
It also introduces an "AI Nutrition Label": a standardized, at-a-glance grade report (A–C style) per model covering negative behaviors it avoids (hallucination, sycophancy, bias, unsafe advice) and positive behaviors it promotes (agency, learning, creativity, wellbeing). A full paper accompanies the release; aimed at AI governance professionals.
More from Safety
- Agent harnesses are crutches; least-privilege action authorization is what matters — andreisavu · 2026-10-09
- Proposed 'comprehension audits': independent auditors grill teams on AI training recipes — ronbodkin · 2026-10-09
- Comprehension audits: gate AI self-improvement on whether humans still understand it — ronbodkin · 2026-10-09
- New Preprint Proposes Comprehension Audits to Gate AI Self-Improvement — ronbodkin · 2026-10-09
- Stephen Casper weighs in on Dario's essay: embedded evaluators, China, regulatory capture — StephenLCasper · 2026-10-09
- Claude Leads 26% of Anthropic R&D; Preprint Proposes Comprehension Audits — ronbodkin · 2026-10-09