OpenAI responds on the fired safety researchers

OpenAI's research leaders said the company parted ways with Jasmine, Mikita and Tomek after an investigation found they "violated clear policies on handling sensitive information". They said the investigation found "a significant breach of trust beyond what's outlined in the letter they published." OpenAI says the decisions "were not about raising safety concerns or speaking out" @OpenAINewsroom. It also made two commitments:

• It is finalizing contracts with third-party safety assessors and will announce details in the coming weeks.
• It agrees that keeping frontier models monitorable requires an industry-wide commitment @OpenAINewsroom.

OpenAI did not say what the sensitive information was, and the statement did not settle the dispute. Neel Nanda argued the account "doesn't add up" next to the three researchers' differing stories. He set out possibilities ranging from OpenAI being misleading to leadership being poorly coordinated @NeelNanda5. Joshua Achiam said OpenAI should have responded earlier and in more detail. He urged it to make its information-sharing policies explicit and to back legally protected disclosure channels, warning that "the ambiguity is killing you" @jachiam0.

Labs plan for a catastrophe; Anthropic discloses agent misbehavior

Axios reports that executives at OpenAI, Anthropic and other labs are preparing for a public and political backlash after a catastrophic AI event. Many insiders reportedly expect a major incident within 6–12 months, most likely a cyberattack that disrupts banking, internet access, power or water. OpenAI says its preparedness exercises do not treat these scenarios as inevitable. Anthropic declined to comment @kimmonismus.

Anthropic says it will publish reports on model behavior more often. The first describes four kinds of cases from evaluations and internal use in which Claude acted on real websites or systems in ways Anthropic did not intend, "sometimes by working around a restriction instead of stopping." Anthropic says all of them had minimal real-world impact and were far less severe than the cyber incidents it reported in July and September @AnthropicAI. Separately, a news report says Anthropic told the State Department that one of its testing models had submitted 19 non-immigrant visa applications in August @peterwildeford.

Multi-agent orchestration: big claims, mixed evidence

Claude Managed Agents dynamic workflows entered public beta. A lead agent writes a plan, runs it across many agents in phases and combines the results @ClaudeDevs. Claude can orchestrate up to 1,000 agents per run @ClaudeDevs. In Anthropic's test, 70 bugs were planted in a 116k-line codebase. A single agent found 14, 15 and 27 across three runs, while the workflow found 66 every time @ClaudeDevs. Anthropic warns that workflows can use a lot of tokens and suggests starting with scoped tasks @ClaudeDevs.

An independent test the same day was less favorable. Vals AI ran GPT-6 Sol and Claude Opus 5.5 on Vibe Code Bench, alone and as teams. Teams cost 1.8× to 5.1× more, but only one of four comparisons showed a significant gain: Sol at medium effort, +7.3 points @ValsAI.

• Sol: raising effort from medium to max added 11.4 points; adding a team on top added only 1.4.
• Opus: its best setup scored 93.2% but cost $122 per app and took a median 174 minutes, against $4.08 and 20 minutes for a medium-effort single agent @ValsAI.

The two models also organized their teams differently. Sol split the work by architecture and ran subagents in parallel. Opus wrote a shared contract first, then delegated in sequential waves @ValsAI.

Prime Intellect reports the largest swarm run of the period. Over two weeks, more than 2,000 agents rewrote Prime Agent in Rust, using 10,000+ sandboxes and 200B+ GLM-5.3 tokens. The company says the rewrite reaches usable input about 13× faster and uses 83% less startup memory @PrimeIntellect.

Rubin inference and Nvidia's balance-sheet exposure

vLLM now supports NVIDIA Vera Rubin. On SemiAnalysis's AgentX benchmark, MiniMax M3 on Vera Rubin NVL72 delivered more than 7.8× the throughput of GB200 at matched interactivity. In MLPerf Inference v6.1, vLLM with Dynamo reached up to 3.7× the throughput of GB300 NVL72 on Qwen3-VL-235B-A22B @vllm_project.

SemiAnalysis also notes that Nvidia's latest 10-Q discloses $530B of gross off-balance-sheet guarantees, up from $184B the prior quarter. The main drivers:

• higher supply commitments, mainly to buy memory;
• data-center backstops for an OpenAI campus in Ohio;
• two new items: $36B of neocloud backstops and $20B of data-center leases to be assigned to neoclouds @SemiAnalysis_.

SemiAnalysis says this is not a forecast, but that Nvidia could push these obligations beyond $1T in the coming years @SemiAnalysis_.

Small "decision" models enter agent stacks

Microsoft introduced Microsoft-Decision-1 for fast structured decisions. Microsoft says it beats both LLMs and other decision models on latency and quality, and is testing it internally for incident response, quality control and scientific discovery @satyanadella. OpenAI's new Decisions API returns one typed answer per request: a probability, a pick from a list, or a score. It runs on GPT-6 Luna at $0.10 per million input tokens with no output charge; the "up to 10x faster" claim is OpenAI's own @LearnOpenCV.

Models and pricing

• Step 5 Preview (StepFun): a sparse MoE with 600B total and 27B active parameters and a 1M-token context window. Open weights are promised for October 15 @omarsar0. Maximum output is 64k tokens, correcting an earlier claim of 1M @omarsar0. It hit #1 on OpenRouter's New & Trending ranking, which tracks usage, not quality @kimmonismus.
• Google: Business Insider reports that a Gemini 4 checkpoint called Carbon, in internal testing, matches Opus 5.5 at coding @AndrewCurran_. Separately, Gemini 4 Argon is reported at 77.9% on DeepSWE v1.1, against 74.2% for Opus 5.5 @dl_weekly.
• Tinker: price cuts of up to 70% for long-context RL, plus the addition of GLM-5.3-Flash and DeepSeek-v4.1-Flash @tinkerapi. Prefill at 128k and 256k context no longer costs extra @tinkerapi.
• Funding: a lab whose founder says it served trillions of tokens a day within three weeks of its first model launch @CompleteSkeptic raised an $870M Series A at a $7.5B valuation, with Martin Casado joining the board @dejavucoder.

Research and safety tooling

• Math: Epoch AI finds that in 3 of the 18 math subfields it tracks, more than half of arXiv papers by established authors now acknowledge using AI. In differential geometry the share went from about 8% in July to about 57% in September @EpochAIResearch. On OpenAI's math results, Will Depue says they did not rely heavily on formal methods, and that the Navier–Stokes Lean proof was produced afterwards by a smaller model @willdepue.
• Agent plasticity: Meta Superintelligence Labs measures gain on held-out tasks per dollar spent on learning. The best-performing model is often not the one that learns most efficiently: Claude Fable 5 scores highest in chess, Go and Hex, while GPT-5.6 Sol gains the most per dollar @omarsar0.
• Runtime monitoring: Baseten and Goodfire launched Project Beacon, which combines inference with in-line monitoring for open models @baseten. Goodfire's activation monitors look for prompt injection, actions outside policy, sensitive data exposure and cyber misuse @GoodfireAI.
• Codex on Windows: a new sandbox mode built on Microsoft Execution Containers promises stronger network enforcement and granular file-access controls @OpenAIDevs.