Safety

MIT Index Finds AI Agents Proliferating With High Autonomy and Thin Safety Disclosures

An MIT index finds widespread autonomy paired with scarce agent‑specific safety evidence.