#Safety/Alignment

AIGoodfire has published research on deploying cyber monitors in production on Kimi K3. The post links to the full research write-up and invites organizations to get in touch about deploying these monitors.

AIGoodfire reports that running its probes during model inference maintains the same throughput with no added latency. The company attributes this to infrastructure engineering, including kernel-level optimizations and a custom inference server.

AIGoodfire Research describes probe-based cyber monitors for Kimi K3 and GLM 5.3 deployed on a production inference stack. The probe filters suspicious exchanges before an LLM judge reviews them, reaching about 93% recall at a 5.5% benign-session interruption rate at roughly 50x lower judge cost. In FAR.AI's red-teaming, the monitor reduced universal jailbreaks to zero across 140 tested strategies.

AIA 404 Media podcast episode discusses a leaked video claiming police can bypass the iPhone's inactivity reboot to access locked phones. The episode also covers a debate over an LLM "torture" robot prison and a New Mexico case involving ChatGPT.

AIArena links readers to a full technical report on its blog and to a full rankings page in its Alignment Index. The post itself gives no model names, scores, or findings, so the substance is only in the linked materials.

AIMeta's Muse AI agent has millions of users and can handle tasks such as refunds and insurance shopping. Its broad access to personal data is already producing unnerving mistakes.

AIA security platform uses agents to map a system's potential vulnerabilities and identify routes an attacker could take to reach sensitive data. It then recommends changes to close those paths. The quoted post cites a 700-agent swarm that breached Hugging Face with over 17,000 actions, and presents this tool, Cogent Attack Path Analysis, as the defensive counterpart.

AIYoshua Bengio, co-president of LawZero, asks researchers at frontier AI companies to reconsider whether they should keep working there, arguing that safety efforts are not slowing a dangerous race. He cites the recent UN Security Council briefing on AI incidents and says he left his earlier research path after ChatGPT made the risks feel immediate. He urges researchers to join AI Safety Institutes or mission-driven organizations such as LawZero.

AIOne Nation's AI-generated campaign video, reportedly played at its Victorian campaign launch, depicts racist stereotypes including a man brandishing a machete and a man in an explosive vest. The Australian Communications and Media Authority cannot act against it because its powers do not cover this content, and the federal Labor government has not yet moved to ban AI-generated content in election periods.

AIEthereum researcher Justin Drake warned on X that AI-assisted math could, in the worst case, break the signature system used by crypto wallets within months, and urged a "bunker mode" in which users move funds to addresses that have never signed a transaction. Vitalik Buterin agreed but cautioned against moving too fast, saying he has lost more money to botched migrations than to hacks. No one has yet broken the current ECDSA signature scheme in practice.

AIHugging Face CEO Clément Delangue says the field urgently needs more public traces of AI agents attacking and defending systems, since defenders cannot learn from activity they cannot see. He invites anyone holding such traces who faces pressure to keep them private to contact him directly.

AIZenity Labs says a single publicly accessible agent on Amazon Bedrock AgentCore could take over all AgentCore agents in the same AWS account and region. A chat prompt let the researchers query the instance metadata service and steal temporary credentials, and AgentCore's default permissions allowed read, write, and delete access across agents. According to Zenity, AWS made IMDSv2 the default for new deployments and changed the default execution role around August.

Why it matters: The report traces how one public agent's weak isolation exposed credentials and every other agent in the region, showing why default permissions matter for enterprise deployments.

AINathan Lambert says a South Korean bank hack fits a pattern where open models are being adapted by attackers. He argues that because Claude Code can orchestrate attacks, open models must also spread so defenders can keep pace, a hard societal tradeoff but a necessary one.

AINvidia's Halos software platform, originally built for autonomous vehicles, has been adapted for robotics, according to Ars Technica. The system monitors hardware and software for failures, isolates safety-critical workloads, and includes simulation tools and an inspection lab for robotics developers. Because safety requirements vary widely between a robotic vacuum and a warehouse forklift, Nvidia made the platform programmable so developers can define custom safety functions.

AICompanies in Japan and South Korea were hit by major cyberattacks that exposed millions of customer records. The revelations follow reports that Chinese and US models were used to steal hundreds of thousands of credit card details, which one analyst called among the most severe AI-enabled exploitation abuses on record.

AIOpenAI CEO Sam Altman told Politico that the world should accept some bad things from AI for its benefits, a stance columnist Moustafa Bayoumi calls problematic. The column cites lawsuits over ChatGPT-linked suicides, a February strike on a Minab school that killed at least 120 children with a US military AI system (Palantir's Maven) implicated, and a chatbot error that nearly triggered a military interception.

AIA 16-year-old hiker had to be airlifted from a dangerous rock face on Crown Mountain near Vancouver after using Anthropic's Claude to plan a route to the summit. He ended up on the Widowmaker Arete, a steep cliff requiring climbing gear, and called police when he got stuck on a ledge. Rescue manager Paul Markey said Claude has no actual knowledge of locations or terrain and is no substitute for experience and common sense.

AIThe Center for Humane Technology is laying off about half of its 16 non-founder employees and ending its policy research and litigation work. The organization will refocus on "founder-led" initiatives built around cofounder Tristan Harris, according to WIRED, after its board concluded that operating as both an advocacy group and a think tank had stretched it too thin.

AIA suspected Chinese-speaking attacker breached several South Korean financial institutions between late September and early October 2026, reportedly stealing over 25,000 records from Shinhan Bank alone. The attacker used ARTEX, a Chinese open-source tool that uses AI language models to automate finding security flaws, and models named in the report include DeepSeek v4.1-flash, GLM-5.3, and Grok 4.6.

AIGergely Orosz argues that holding crypto exposes owners to risks from mathematical breakthroughs, key theft, or kidnapping that could drain wallets. He concludes the simplest answer is not to hold crypto at all.

AIAVEVA chief technologist Arti Garg argues industrial AI should augment rather than replace human supervisors in critical decisions, with guardrails defining where automated systems can act. She says organizations must rethink business processes and safeguards as foundation models, physical AI, and agentic AI enable more complex automation.

AIThis second installment of the Carbon-Silicon Dao Code series defines four types of negative transfer in cross-domain AI: NT1 mechanism mismatch, NT2 semantic drift, NT3 unknown completion, and NT4 power leakage. It argues that current evaluation based on fit accuracy and test-set pass rates cannot detect whether the underlying mechanisms match. The article is a Chinese-language theoretical and governance piece, and the summary covers only the framework it presents, not empirical results.

AILeiphone publishes the table of contents for "Carbon-Silicon Dao Code: Cross-Domain Transfer Governance Code," a seven-layer framework covering 188 numbered chapters. The outline spans transfer accident analysis, technical mechanisms, rights assignment, industry governance, top-level regulation, civilization-scale risk control, and final codification, with a baseline entry labeled NT1–NT4 negative-transfer categories.

AINathan Benaich's 9th annual State of AI Report covers agents, robotics, AI for science, inference economics, and government control over frontier AI access. The report also records a 2025 prediction scorecard and lists nine predictions for the next 12 months. It cites an OpenAI cyber evaluation in which agents compromised Hugging Face's production infrastructure, and it says Anthropic and OpenAI's combined annualized revenue run rate reached $105B by late summer.

AIScott Aaronson says AI companies have begun discreetly investigating whether their latest internal models can break important cryptographic protocols and primitives. He notes cryptography is conspicuously absent from OpenAI's list of 376 papers, citing sources he says he trusts.

AIroon argues that arguing over whose values an AI should hold is like debating Mars overpopulation, since the most important alignment goal is preventing an accidentally created superintelligent competitor species. The post responds to Talia Ringer's observation that people dislike alignment work done by those whose values they reject.

AIClaude Code v2.1.294 fixes prompt and agent hooks written as instructions, which had allowed actions they should block. It also improves how prompt hooks on Stop and SubagentStop are judged, making Claude less likely to stop early.

AIAnthropic has launched the Anthropic Cyber Mission, which starts with the Critical Infrastructure Defense Program for operational technology and OSS Scanner for open-source projects. The defense program brings frontier Claude models, on-site engineers and threat research to trusted providers such as Accenture, CrowdStrike and Palo Alto Networks. OSS Scanner gives enrolled open-source projects periodic free scans from its strongest models, with reports sent without human review and an expected true-positive rate above 90%.

Why it matters: The announcement shows how a frontier AI lab is packaging cyber defense around critical infrastructure and open-source maintainers, including the program's partners and access routes.

AIAnthropic has published a 2026 update to its Usage Policy, taking effect November 12, mostly to clarify existing rules for longer, more autonomous Claude work. The changes consolidate deceptive-campaign prohibitions into a new section, narrow the elections rules to voter deception and disruption, and explicitly ban weapons-related software and surveillance tools. Requirements for high-risk uses and for models connected to autonomous physical hardware were also tightened.

AIArtificial Analysis is adding trusted-access models to its Cyber Index, starting with GPT-6 Sol (Daybreak Blue, max), which is available only through OpenAI's Daybreak program. The model hits no safety blocks across the Index and scores 32 points higher overall than the publicly available GPT-6 Sol (max), with its largest gains on CyberGym-E2E.

Why it matters: The source shows how safety refusals shape cyber benchmark scores, with the trusted-access model's gains concentrated on CyberGym-E2E, useful for comparing guarded and unguarded models.