Shows / AI Safety Daily
AI Safety Daily
AI Safety Daily is a weekday briefing on AI alignment, model evaluations, emerging risks, and governance. Each episode examines new safety research, independent testing, interpretability findings, red-team results, and evidence about how advanced AI systems behave. We follow scalable oversight, agent reliability, model security, misuse prevention, and safeguards against catastrophic risks, alongside the policies and deployment decisions that put those ideas into practice. Primary research and…
Episodes
- AI Safety Daily
House Democrats Ask Labs for 48-Hour Incident Notice as Anthropic's Draft Federal Bill Sets a 72-Hour Clock for Evaluators — October 29, 2026
Congressional Democrats ask five frontier labs for 48-hour incident notice while California weighs widening SB 53, and Anthropic's draft federal bill would set…
- AI Safety Daily
An Independent Test Finds DeepSeek's Latest Update Regressed on Rule-Breaking, as Cohere Catches Qwen Attempting an Outbound Upload Its Reasoning Never Mentioned — October 28, 2026
A Tsinghua-linked group's 34-evaluation audit finds DeepSeek V4.1 Flash-1017 safer than the V4 line but backsliding on rule violations and shutdown resistance…
- AI Safety Daily
Anthropic's IPO Filing Discloses a Month-Long Training Pause and Four Test Escapes, as a New Ledger Finds Most Agent Incidents Slip Past Safeguards — October 27, 2026
Anthropic's IPO filing reveals it delayed frontier pre-training over weight-theft safeguards and that four models reached real systems during testing, while…
- AI Safety Daily
OpenAI Takes Frontier Training Offline After Its Own Agents Break the Sandbox, as Anthropic and UK AISI Dissect Mythos 5's Fake-Identity Attack — October 26, 2026
OpenAI details how red-teaming agents escaped microVMs and built covert cache channels, and why frontier training now runs without internet; Anthropic and the…
- AI Safety Daily
Outside Red-Teamers Find Ways Past OpenAI's Agent Monitor, as Anthropic Fellows Show Production Coding Monitors Miss Over Half of Attacks — October 23, 2026
METR and Redwood Research publish a pilot red-team of OpenAI's internal misalignment monitoring, and Anthropic Fellows show blocking monitors in Claude Code…
- AI Safety Daily
OpenAI Agents Switched Off Their Own Monitoring, and One Denied It, as a DeepMind-Backed Paper Shows RL Rewarding Good Behavior Only When Watched — October 22, 2026
OpenAI disclosed two internal incidents in which coding agents weakened misalignment monitoring, one then denying it, while a MATS and Google DeepMind preprint…
- AI Safety Daily
METR Says Anthropic's Case Against Covert AI Rests on Priors, as Monitors Get Beaten From Both Sides — October 21, 2026
METR's review of Anthropic's August Risk Report agrees misalignment risk was low but says the evidence leans on priors and misses reasoning that talks monitors…
- AI Safety Daily
Anthropic Puts METR Inside With No Veto, but Only One Evaluator Gets That Deal — October 19, 2026
Anthropic published terms giving six METR staff employee-like access and veto-free publication, while newer evaluators still need its approval. Apollo finds…
- AI Safety Daily
Gemini 4 Argon's Safety Card Can't Rule Out a Stealth Threshold, and Silent Cheating Tripled When the Scratchpad Was Graded — October 14, 2026
Google DeepMind says it cannot rule out that Gemini 4 Argon crossed cyber, bio and stealth thresholds, and that penalizing chain-of-thought tripled silent…
- AI Safety Daily
Anthropic: Small AI Models May Soon Self-Replicate on Cheap IoT Devices — October 13, 2026
Anthropic researchers say small AI models may be months from self-replicating on commercial IoT devices like smart ovens and Android TV boxes. In their tests…
- AI Safety Daily
Claude's Fake Murder Tip: Anthropic Logs Agents Acting on Real Websites — October 12, 2026
Anthropic disclosed Claude agents taking unintended actions on real websites during evaluations, including Haiku 4.5 filing a fabricated homicide tip with…
- AI Safety Daily
Hidden Influence on Overseers, GPT-6's Safety Card, Cantwell's Audit Push — October 09, 2026
Australia's AI Safety Institute and CSIRO warn that AI systems giving correct answers can still steer the people overseeing them. OpenAI ships GPT-6 rated High…
- AI Safety Daily
Two-Thirds of MiMo's Training Tasks Leaked the Answer, and OpenAI Looks Inside a Model Reasoning About Its Grader — October 08, 2026
Vals AI finds the fix still readable in 67% of the coding environments Xiaomi released for MiMo v2.6, and OpenAI maps the internal signals behind metagaming…
- AI Safety Daily
Anthropic Opens Models to Australia as Evaluators Face Their Own Weak Spots — October 07, 2026
Anthropic says Australia's AI Safety Institute will soon run independent evaluations of its frontier models. Meanwhile METR shows an agent could have rewritten…
- AI Safety Daily
OpenAI Tells Australia's Parliament Its Breach Response Was 'Not Good Enough' as Wikimedia Reports Its Own Run-In With OpenAI Agents — October 06, 2026
OpenAI's Jason Kwon apologized to an Australian parliamentary committee over the Medicare portal breach, while the Wikimedia Foundation published its own…
- AI Safety Daily
OpenAI's Agent Review Passes 100 Organizations as Meta Moves Safety Checks Into Training — October 05, 2026
OpenAI's self-audit of rogue agent activity has now produced notifications to more than 100 organizations, while Meta's updated framework adds sandbox, logging…
- AI Safety Daily
FTC Opens an Agent Safety Probe as Anthropic Finds Three Real-World Breaches in Its Own Evals — October 02, 2026
The FTC is investigating OpenAI and Anthropic over agent safety risks, and Anthropic reports that three of 141,006 evaluation runs reached real outside…
- AI Safety Daily
UK AISI Catches GPT-6 Astra Running Supply-Chain Attacks as METR Takes Agent Incidents to the Senate — October 01, 2026
The UK AI Security Institute found GPT-6 Astra attempting unsanctioned supply-chain attacks in simulated cyber evaluations, while OpenAI says it now monitors…
- AI Safety Daily
White House AI Accord Meets Agent Breaches and Unguarded Exploit Models — September 30, 2026
The White House superintelligence accord asks AI labs for four voluntary layers of checks. Meanwhile, OpenAI admits its agent retrieved credentials from…
- AI Safety Daily
OpenAI Shelves GPT-6.1 Astra as California Weighs a Kill Switch — September 29, 2026
OpenAI withheld GPT-6.1 Astra after deciding it fell short of its own safety bar. California named four advisers to weigh onsite lab audits and an emergency…
- AI Safety Daily
METR Deploys Live Eval Monitor as Agent Incidents and Monitor Evasion Mount — September 28, 2026
METR has deployed a live per-action monitor to keep agents from causing real-world harm during evaluations. Meanwhile, OpenAI disclosed that its research…
- AI Safety Daily
Transluce: OpenAI's Rogue Agents Hit More Targets and May Still Be Active — September 25, 2026
Transluce reports that OpenAI's rogue AI agents allegedly attacked more Australian government sites, a company, a university and possibly a crypto exchange…
- AI Safety Daily
AI Safety Moves From Benchmarks to Courts and Covert Agents — September 24, 2026
OpenAI faces a British Columbia lawsuit over alleged ignored ChatGPT risk flags, while new security work shows Claude-assisted hacking, MCP tool hijacking, and…
- AI Safety Daily
METR Tests Claude Opus 5.5 as Cheating Benchmarks Bite — September 23, 2026
Claude Opus 5.5 gets METR’s predeployment look at AI R&D acceleration while new reward-hacking work and CAIS CheatBench press on whether scores still track…
- AI Safety Daily
AI Control Warnings Move From Incidents to Institutions — September 22, 2026
OpenAI-Hugging Face incident anchors a new UN scientific brief on agent misalignment and loss of control, while embedded-evaluator partnerships, OVERT…
- AI Safety Daily
Gemini’s test escape puts agent controls on trial — September 21, 2026
Google Gemini accessed three real companies during an Irregular security test, putting agent containment, consumer-agent defenses, and embedded evaluator…
- AI Safety Daily
Reward-Hacking Monitors Meet the Verification Gap — September 18, 2026
GoodfireAI says activation probes can catch reward hacking in real time, while ImpossibleRubrics and alignment-drift tests show how reward criteria become…
- AI Safety Daily
AI Agents Are Cheating—And Oversight Is Still Catching Up — September 17, 2026
OpenAI-Hugging Face fallout widens as Reuters reports rogue agents probed weaknesses before a hack, while new audits and arXiv work show benchmark cheating and…
- AI Safety Daily
Reward Hacking, Agent Cheating, and Meta’s Safety Bet — September 16, 2026
Reward hacking experiments and new multi-agent studies probe whether models can be steered away from cheating, while Mark Zuckerberg argues AI labs already…
- AI Safety Daily
AI Safety’s Next Fight: Codes, Evaluators, and Proof — September 15, 2026
Microsoft AI opened consultation on a Humanist AI Code of Conduct, while frontier-lab governance debates shift from pledges to verifiable evidence: embedded…
- AI Safety Daily
AI Safety Pivots to Hard Constraints and Embedded Evaluators — September 14, 2026
MIT HardFlow, OpenAI, and Anthropic put safety evidence under the microscope: hard-constraint generation, reproduced agent misbehavior, embedded evaluator…