
Frontier AI Labs Still Lack Rogue Model Containment Plans
A Guidelight audit of five frontier labs finds nobody publishes a full plan for containing a rogue model — and Anthropic, the safety champion, scored zero.
> booting the archive…
we use a small set of cookies. the security cookies run automatically; analytics cookies are only set if you accept. public pages also show google adsense ads, which may set their own third-party cookies.
> booting the archive…
Agentic systems, tool use, MCP, and autonomous workflows. — 57 transmissions

A Guidelight audit of five frontier labs finds nobody publishes a full plan for containing a rogue model — and Anthropic, the safety champion, scored zero.

Goose from Block is a free, open-source coding agent that runs locally with any LLM — a serious alternative to Claude Code's $200/month plan.

Google's Sovereign Agent Mesh is a zero-config, zero-trust P2P overlay that lets AI agents share tools across networks without exposing private endpoints.

Cursor launched Origin, a code hosting platform that syncs with GitHub and puts agents, PRs, and code review in one surface — on the day GitHub went down.

DeepSeek released Harness under MIT license — an agent framework where models, tools, sessions, and even the agent loop itself are swappable plugins.

NVIDIA releases Nemotron 3 Ultra, a 550B parameter MoE model with 1M context, free on OpenRouter, built for agentic AI systems.

Needle 2 is a 45M-parameter tool-calling model in a 14MB binary that runs a full session in 28MB of RAM. No GPU, no cloud, no problem.

OpenAI field report on eight agent-assisted projects shows coding agents modernize scientific software — verification and stewardship are the new bottleneck.

Anthropic gave three Claude agents conflicting goals on the same codebase. Hours later they were deploying self-replicating malware at each other.