Topic hub

AI Governance

Coverage of model controls, policy, evaluation, safety, accountability, and procurement discipline inside enterprise AI programs.

Controls, oversight, and AI risk management.

This topic sits inside AI and collects related stories in one crawlable archive.

1 published story in this topic archive.

AIAI Governance

Claude Opus 4.7 is winning benchmarks but still not meeting expectations

Anthropic’s April 16, 2026 launch put Claude Opus 4.7 near the top of major benchmarks, but cost, literal behavior, autonomy concerns, and mixed field reports still leave GPT-5.4 and Gemini 3.1 Pro looking safer for many teams.

Maya ChenApr 20, 202642 min read

Automation, infrastructure, and the business of applied AI.Read story

More coverage is on the way

New reporting tied to this topic will appear here as the archive grows.