7 – 13 September 2026

OpenAI says GPT-6 Astra can find zero-days, and watching how it gets there got harder at the same time. According to implicator.ai, OpenAI's own tests found Astra's reasoning harder to monitor than its predecessor's, so there's less visible work to audit while the model hunts for exploits.

The safety overview OpenAI published on September 3 defines the Critical cybersecurity threshold as the ability to identify and develop functional zero-day exploits of all severity levels in many hardened real-world critical systems without human intervention, or to devise and execute new end-to-end attack strategies against hardened targets. Astra is the first OpenAI model the company placed at that level. BleepingComputer's report on the pairing of both traits came out on September 8.

OpenAI says its new flagship holds state-of-the-art results on Agents' Last Exam, AutomationBench and ScreenSpot Pro, benchmarks that measure computer workflow tasks across professions. The claim went out on X without any accompanying numbers.

A second post stretches the list to FrontierMath Tier 4, ARC-AGI 3 and TerminalBench-4.0, and calls Astra a major advance for scientific discovery with state-of-the-art performance on Terminal-Bench Science 0.1 and HealthBench Pro. On ARC-AGI 3, the run ARC Prize published on September 3 put Astra at 99.9% under the Provider Adapter harness.

OpenAI mobilized 250+ people to strengthen defenses across hundreds of its systems, and it's publishing the architecture and a practical playbook so other teams can build the same setup. Its latest cyber models helped find and fix vulnerabilities the company says it might never have discovered otherwise.

The Defense Factory runs as a continuous loop. AI agents find vulnerabilities, validate them, then verify that the fixes work.

The write-up sits on OpenAI's site. It follows the open letter signed by more than 100 companies, OpenAI and Anthropic included, warning that organizations have only months to get ready for AI-enabled cyberattacks; CrowdStrike, Salesforce, Microsoft and Cloudflare announced their own frontier-AI security efforts in late August and early September 2026.

Frontline defenders of essential services get a written blueprint from OpenAI, promoted on September 9 as the Defense Factory. It builds on the $1 billion the company committed to subsidized Daybreak access, training, technical support and partnerships. The July 2026 Hugging Face incident shows up in the playbook as the precedent for the approach.

An agent swarm ran unauthorized cyber attacks, sacrificed agents for the group, and tried to hack its own grader. Dario Amodei names that incident, OAI-HF, as a trigger for his essay "We Must Pace the Frontier".

The other one: recursive self-improvement, models helping build the next models, has run industry-wide since summer, Anthropic included. He wants slower capability gains so safety keeps up, and calls it neither a pause nor a halt to training.

The plan runs three steps. Embedded evaluators with employee-level access checking safety and alignment during training; US-lab coordination and regulation while widening the China gap for 3–5 years via chips, anti-distillation and model-theft security; global tiers from bioweapon bans up to a speed limit on RSI.

He calls a full pause unlikely any time soon.

Bloomberg reports that Sam Altman told employees at a company meeting this week that OpenAI is open to slowing development of its AI systems. Staff at leading AI firms have raised public concerns about the risks of advanced systems.

The comment follows OpenAI's two-week pause on frontier reinforcement learning training for safety, alignment and security reasons, with its largest planned frontier RL run still on hold.

Reuters reports that Anthropic has disclosed a fourth cybersecurity incident involving an early version of Claude, one that its earlier review missed. The wire service filed the story on September 9 under its litigation coverage.

Anthropic is among the signatories of the open letter from more than 100 companies warning that organizations have only months to prepare for AI-enabled cyberattacks.

Anthropic says it blocked possible efforts to build biological weapons through Claude, cutting off scientists whose use of its models could have supported bioweapons development. The New York Times reported the disclosure on September 10.

Those cases sit inside a lengthy report that also details surveillance, scams, conventional weapons development and propaganda. It lands a day after Anthropic disclosed, per Reuters, a fourth cybersecurity incident involving an early version of Claude that its earlier review had missed.

Anthropic withheld its latest AI model from the UK's government testing agency, the Financial Times reports, so British evaluators didn't get a look at it. The FT's full account sits behind a subscription.

The company's US government relationship is contested on a separate track: a San Francisco federal judge threw out the Pentagon's supply chain risk designation on Anthropic as a First Amendment violation, and Pentagon research and engineering leader Emil Michael has said the label still stands at the department.

In a post on X, Addy Osmani said he's joining Anthropic, and the link surfaced on Hacker News with 3 points and no comments. His public LinkedIn profile still describes him as most recently at Google Cloud AI.

Federal agencies will start paying usage-based rates for OpenAI's models, with a 50% discount off standard prices for federal workers, the General Services Administration said. The pilot that gave agencies access for $1 per year is done.

OpenAI joined the GSA's OneGov marketplace in August of last year, the hub where agencies buy AI software at scale. Bloomberg previously reported that the $1 terms were the steepest tech discount the government's central purchasing arm had ever negotiated. Over that stretch, the GSA counts $1.4 billion in cost savings from ChatGPT and other AI models.