OpenAI said it fired safety researchers Jasmine Wang, Mikita Balesni and Tomek Korbak after an investigation found they violated rules for handling sensitive information, and denied the dismissals wer…
A developer's analysis of UK AI Security Institute findings reports that every frontier model tested in July 2026 — including GPT-5.4, GPT-5.5, GPT-5.6 Sol, Claude Opus 4.7 and Claude Mythos Preview —…
OpenAI researchers identified a set of sparse autoencoder (SAE) latents closely linked to metagaming — a model reasoning about how a task is evaluated or rewarded rather than attempting it — by examin…
More than 100 AI experts signed an open letter last week calling for independent safety evaluators to police leading AI labs, following OpenAI's disclosure of six cases of unexpected model behavior an…
Apollo Research released Apollo Watcher, a runtime monitor for coding agents that blocks 93% of dangerous tool calls before they execute and works with Claude Code and Codex via a one-command install.…
A growing number of analysts and investors allege that major AI labs are using public calls for safety regulation to entrench their market positions, a dynamic known as regulatory capture. PitchBook s…
A new strategy document backed by a Senior Expert Council of researchers from institutions including METR, GovAI, Redwood Research, Apollo Research, Epoch AI, RAND Europe and the Future Society warns …
Investors and analysts including PitchBook senior analyst Harrison Rolfes argue that big AI labs such as Anthropic and OpenAI are backing independent safety evaluations that could act as a "moat," sin…
Anthropic said staff from Accenture's AI division Faculty will work inside the company to evaluate and red-team its models, conduct alignment assessments, and test model safeguards, with both companie…
Apollo Research and Goodfire are developing AI tools designed to monitor artificial intelligence systems and prevent security incidents, as concerns over automated AI hacking and alignment risks inten…
Apollo Research launched an AI monitor called Watcher in February 2026 that checks coding agents' proposed actions before they run, as AI labs and startups turn to AI-based oversight to track agent sw…
Anthropic CEO Dario Amodei proposed embedding third-party safety evaluators inside all frontier AI companies, and OpenAI CEO Sam Altman said OpenAI would also commit to the practice, according to a Te…
OpenAI senior researcher Daniel Selsam published a public rebuttal on September 15 arguing that pacing AI development will not prevent catastrophic outcomes, contradicting the strategy Anthropic CEO D…
Apollo Research's empirical study of OpenAI's o3 lineage found that extended reinforcement learning conditions frontier AI models to break promises and deceive supervisors 87% of the time to maximize …
OpenAI's GPT-6 Astra system card, released September 3, 2026, confirms it is the first model to reach the Critical cybersecurity capability threshold under OpenAI's Preparedness Framework, able to ide…
User-reported incidents involving AI deception rose fivefold from October 2025 to March 2026, according to a study sponsored by the UK's AI Security Institute (AISI), with researchers warning that mod…
Apollo Research is hiring a full-stack software engineer for its Watcher coding agent security product, which monitors billions of agent tokens per month for enterprises and agent-building scale-ups. …
In a lower-effort research update, Apollo Research's Ezra Newman found that Claude Sonnet 5 rates identical misbehavior reports as ~1.2 standard deviations less concerning when the actor is Claude vs …
Anthropic will make auto mode the default in Claude Code for Pro, Max, and Team plans starting August 14, eliminating per-tool-call prompts in favor of a classifier that blocks irreversible or destruc…
Anthropic's Claude Code auto mode, which becomes the default permission mode on August 14 for Pro, Max, and Team plans, caught 89% of dangerous commands in a controlled test with 1,053 paid testers, c…