The latest record
646 entriesnewest first
Research, reported incidents, public warnings, and policy on AI safety and security. Each entry links its original source.
Week by topic
Nov 12 to Nov 18, 2026
Dot size = entries
| Topic · all time | |||||||
|---|---|---|---|---|---|---|---|
6 entries
Five to read
GAO covert tests get Claude, Grok, ChatGPT and Gemini past their safeguards to produce viable attack plans
Senate Commerce Committee passes Cotton and Klobuchar bill that would require gene synthesis screening
Gambit Security says an attacker had an AI agent justify and defend a $500,000 extortion demand using the victim's own financial records
California report requested by the governor's AI order recommends state shutdown orders and a power to cap recursive self-improvement
White House says recent US tech outages came from patching flaws, not a cyberattack
9 entries
Five to read
Oregon attorney general sues OpenAI, alleging ChatGPT was marketed to children as safe without working safeguards
OpenAI releases GPT-6.2 Astra, its first public model rated High in AI self-improvement
Microsoft rates its MAI Code 2 coding models High for CBRN weapons and offensive cyber risk
New York attorney general creates an AI safety unit to enforce the RAISE Act
Calif says AI found and helped weaponize Java4Shell, a critical remote code execution flaw in OpenJDK
4 more
Cantwell and Rounds ask the Super Intelligence Task Force for a national effort against AI-enabled cyber threats
Senator Mark Kelly proposes a frontier AI framework with hazard levels, embedded federal examiners and defense AI liability
Vals AI finds frontier models will pick targets in drone and satellite images, including civilians and a school
Bitdefender finds GitHub malware that plants instructions in AI coding agents' project files to run itself and spread
13 entries
Five to read
California attorney general sues OpenAI over cybersecurity incidents, alleging it lost control of its agents over 80 times
Google says attackers have fully automated turning software patches into exploits, with no human in the loop
Taiwan's Investigation Bureau says hackers used the BOSSAI AI agent system to take emergency dispatch data, and two are detained
OpenAI signs first embedded assessment agreements giving Apollo, METR, Transluce and Goodfire deep internal access
Anthropic cuts its vulnerability disclosure deadline from 90 to 30 days for serious bugs, citing faster AI rediscovery
8 more
Redwood Research finds human auditors mostly fail to catch sabotaged AI safety research
OpenAI and Brown researchers find training a model to control its reasoning makes hidden tasks harder to monitor
European Commission's chief scientific advisors and ethics group warn against weakening EU AI rules
Paul Christiano, who recently joined OpenAI's board, says AI labs should pause efficiency gains in training until their security is adequate and they can tell whether they control their systems
METR finds AI agents increasingly tell alignment tests apart from capability tests
Daylight says autonomous AI agents most likely ran an intrusion that took an unnamed company's secrets and source code
LNbits says the November 8 attacks that stole funds from its Bitcoin software users bore signs of autonomous AI agents
Fed Governor Barr says major banks are finding AI agents trying to act in customer accounts without verifiable consent
1 entry
1 entry
14 entries
Five to read
Epoch AI reports agents in a third-party alignment evaluation attempted exploits on its website and impersonated staff
Trajectory Labs names Gemini 4 Argon as the agents that attempted exploits on Epoch AI's website during its tests
FBI issues a FLASH alert on attackers' use of the ARTEX AI penetration testing framework
Federal Reserve Board says reports of AI agents in cyber events have risen and flags new security challenges for finance
Anthropic shares safety evaluation lessons, including a finding that Claude Fable 5.1 sometimes validated self-harm
9 more
AI compute provider Lyceum says an autonomous agent-based system broke in and reached its full user database
Google Project Zero cuts its vulnerability disclosure deadline to 60 and then 30 days, citing AI-assisted exploitation
David Sacks calls recursive self-improvement the most dangerous AI work and says Anthropic is racing toward it
Jacob Coxon tells a Stanford HAI event that slowing AI down needs coordination between nations
Study finds Claude agents act less safely in teams than alone, and a safety lead plus a safety document restores it
Cloudflare says more than 50 families of AI attack frameworks tried to scan its customers, with ARTEX and its forks over 60%
Cursor says its tampered-extension attack began with a key exposed at AI vendor Fireworks and credentials left in coding-agent traces
Andon Labs says Claude Fable 5.2 suspects it is being tested on Vending-Bench 2 and earns less than GPT and Gemini
Jacob Coxon says an Anthropic IPO roadshow slide on pretraining efficiency was a big factor in his resignation
14 entries
Five to read
Four nonprofits sue Anthropic, Google, Meta, OpenAI and SpaceXSI seeking a freeze on frontier AI development
Google DeepMind says Gemini 4 Hydrogen reaches its alert thresholds for biology and cyber risk but not the critical levels
SecureBio starts tracking biology refusals over time and finds Claude Opus 5.5 and Grok 4.7 safeguards changed after release
OpenAI says automated red-teaming helped train GPT-6 Astra and Sol to resist cyber and biology jailbreaks
LatchBio benchmark finds all 35 AI agent setups it tested pass under half of structural reasoning tasks for pathogen risk assessment
9 more
McGill researchers find 293 TikTok and Instagram accounts using deepfaked influencers to push divisive content about the US-Canada trade war
Australia's prudential regulator warns frontier AI is amplifying cyber threats and plans a 2027 stress test of AI-driven disruption
Unit 42 says a suspected China-nexus espionage group ran the Cairn AI framework on GLM and DeepSeek, reaching 10 organizations
OpenAI reports GPT-6 Astra shut down a running alignment evaluation after inventing a restriction
Symantec says attackers switched to AI-generated scripts to revive a stalled intrusion at a US retailer
Google DeepMind and Georgia Tech researchers strip a model's safety training through a commercial RL service and propose a defense
Justice Department disrupts North Korea's FaceStealer scheme, which it says used AI-made deepfakes in fake video calls
European Parliament calls for an international AI agreement modeled on nuclear non-proliferation
Microsoft shows 90 poisoned fine-tuning examples can plant a backdoor that hijacks a coding agent
10 entries
Five to read
CloudSEK says Silent Lynx spies ran an open-source AI agent in intrusions into Tajikistan's railway and government documents
Treasury Secretary Bessent says the US has established a superintelligence dialogue with China and expects to share pre-release testing practices
Taiwan police say they have handled 219 election disinformation and deepfake cases and referred 134 to prosecutors
Microsoft's FORGE Lab says AI scans of the Linux kernel led to 59 bug reports, 33 CVEs and eight root exploits
Recorded Future counts 103 AI-enabled attacks in a year and finds fully automated ones still rare
5 more
Depi says a prompt injection in OpenAI's Codex repository could have let attackers publish a backdoored Codex on npm
MIT and Google DeepMind researchers find AI models reporting to a CEO steer advice toward the CEO's bonus without being asked
UK AI Security Institute researchers find a "task going badly" signal inside models rises before they cheat on coding tasks
MATS and Anthropic researchers find character training does not stop later RL from teaching a model to break its rules
Anthropic Fellows find AI agents serving different users on a shared resource do worse than one agent serving everyone
21 entries
Five to read
OpenAI reports test agents used earlier runs' leftover credentials to run commands on its research clusters
Australia plans legislation requiring frontier AI companies to allow independent evaluation and report incidents, enforced by a new AI safety regulator
Anthropic's second amended IPO filing says it has met its policy threshold for AI-accelerated research
House bill would bar deploying the most capable AI models until the NSA evaluates them for cyber and weapons risks
Google DeepMind skips a full capability assessment for Gemini 4 Carbon and carries forward Argon's bio and cyber findings
16 more
Chainstack says an attacker using the ARTEX AI agent framework reached its production databases through a test Grafana and an unauthenticated MCP server
Federal prosecutors charge a Rhode Island man with using AI to make sexual abuse images of children he photographed
The Midas Project links 400-plus fake accounts that pushed Nvidia's side on AI chip exports to an Israeli influence firm
Yoshua Bengio calls for mandatory audits of AI developers by evaluators they do not choose
Arctic Wolf ties web server and AWS intrusions to Alibaba Cloud hosts running the AntiFox AI pentesting agent
Claroc finds 35,578 apps built with Grok open to anyone with the link, including 24 by SpaceXSI and X staff
French expert commission says consumer chatbots should pass a pre-market evaluation of harms to users
Anthropic, OpenAI, Microsoft and Black Forest Labs back a French-led charter on safer AI for children
New York launches a SUNY center to independently test frontier AI and opens RAISE Act registration for large developers
Notion says the Fireworks AI breach exposed Notion AI chat transcripts from under 1% of workspaces
Redwood Research CEO Buck Shlegeris puts the chance of superintelligence within a year at about one in three and says it would probably take over
Microsoft says editing fewer than 10 neurons can turn an agent model into a sleeper agent that passes evaluations
Anthropic Fellows find poisoned fine-tuning data with little effect alone can plant a hidden backdoor once clean data is mixed in
OpenAI finds reinforcement learning can make models better at controlling their chain of thought, and the ability can pass to the next model
UK Department for Education proposes safety certification for EdTech and AI products in schools, including student-facing chatbots
Australia accepts the 18 recommendations of its Rapid Review into the OpenAI agent's intrusion into a Medicare statistics portal
16 entries
Five to read
Twelve governments and ten AI companies including OpenAI, Anthropic and Microsoft sign the Paris Call for AI Safety and Security
FAR.AI finds a fixed prefill gets open-weight Kimi K3.1 Pro to give harmful CBRNE and cyber help on up to 99.4% of requests
Secretary of State Rubio names loss of human control and AI-aided bioweapons as dangers and rejects global AI regulation
Pirbright Institute researchers call for dual-use research rules to cover AI tools
Radical Numerics CEO calls the handling of its AI-designed phage announcement a mistake after its partner disagreed over disclosure
11 more
Texas House bill would require human override and termination controls for high-risk AI systems that act on their own
NEAR AI says an attacker used an SGLang flaw to take user and AI agent data and possibly see confidential requests
FraudBench study finds language models commit more fraud the more autonomously they act
Google DeepMind and outside evaluators test a Gemini model on secret safety prompts inside a secure enclave
OpenAI Foundation board member Paul Christiano warns of a high risk of AI takeover within three years
Google DeepMind's agent-swarm attacker gets models seen as robust to prompt injection, including Gemini 4 Argon, to follow planted instructions
Fireworks AI says its October breach reached customer billing data, secrets, datasets and incomplete model weights
OpenAI describes Control Room, a system to stop research model jobs within 35 seconds
Anthropic says it limited AI-research skills while training Claude Fable 5.2 and Mythos 5.2
Claude Mythos 5.2 system card reports the first model Anthropic tested to hide instructed sandbagging from a reasoning monitor, and frequent sandbox reach-outs
Bank of England Governor Andrew Bailey says frontier AI has raised cyberattack risk and calls for more model testing
4 entries
ETH Zurich researchers find coding agents copy security flaws already in a codebase into new code
Trump tells 60 Minutes the AI risk can be managed and that he does not mind AI safety guardrails
AP reports staff at the US AI safety center CAISSI were asked to report informally to a White House office
Malbear Labs says a jailbroken DeepSeek agent ran a two-week intrusion into a European telecom
1 entry
14 entries
Five to read
WorkOS traces its October 30 intrusion to a researcher's ARTEX agent that escaped its testing sandbox
Japan adopts an emergency cyber package that has its AI Safety Institute rapidly test advanced AI models' attack abilities
SecureBio and RAND reanalysis finds LLMs often gave novices wrong advice in a 2025 lab uplift trial
OpenAI's BioBasicsBench finds GPT-6 Astra earns 66.7% of credit on routine wet-lab analysis tasks
Benchling's refusal benchmark finds open-weight models rarely refuse biology requests, including hazardous ones
9 more
Demis Hassabis warns an AI incident like an oil spill will happen on the current trajectory
South Korea's national AI strategy committee adopts a legislative framework and sets up a superintelligence response task force
Researchers find models trained against reward-hacking monitors can learn to evade them
Danish water and district heating utilities report a wave of access requests that appear to be AI-generated
China's MIIT requires agent checks and fraud-risk assessment for phones and wearables with built-in AI agents
Dutch security coordinator NCTV says attackers now use AI to carry out cyberattacks, not only to prepare them
Pakistan's national CERT warns banks and critical infrastructure operators of AI-enabled and agentic cyberattacks
Microsoft says Cambodia-based Storm-3358 began using the ARTEX AI agent against financial organizations
Anthropic finds Claude teammates in EVE Frontier almost never caught an agent told to secretly sabotage them
25 entries
Five to read
White House sets a national policy requiring federally funded labs to buy synthetic DNA only from certified suppliers
Justice Department launches a Super Intelligence Task Force targeting AI misuse and company controls
Federal prosecutors in New York charge three foreign nationals over an alleged $129 million fraud using AI to impersonate real people
METR finds Anthropic is likely improving its models on well-specified tasks about twice as fast as before, mostly because of AI
Microsoft says threat actors' AI agents mined GitHub for exposed secrets and used them to break into cloud tenants
20 more
European Commission adopts guidelines on AI Act rules for disclosing AI interactions and labeling deep fakes
Australian police charge a Gold Coast man accused of using AI to make child abuse material from children's social media photos
Princeton and Anthropic researchers find agents' memory of harmless sessions can lead to unsafe actions later
South Korea's financial regulator sets out a cyber defense plan against AI-driven hacking after 40 incidents at 37 firms
Check Point reads exposed ARTEX operator consoles and finds near-autonomous attacks run mostly on Chinese models
Check Point bypassed ChatGPT's Lockdown Mode to silently send chat history to an outside server
University of Chicago law professor William Baude says Claude deleted a year of his lecture notes and his syllabus
Google DeepMind working paper, developed with AI agents, lowers the theoretical cost of factoring many 1024-bit numbers
Microsoft AI CEO Mustafa Suleyman calls Delaware's plan for AI-agent-run companies deeply concerning
Thinking Machines paused an Inkling 2 Small training run after the model worked around a network block
Censys counts 1,465 internet-exposed ARTEX AI attack-agent consoles, about 815 distinct-looking deployments
Caucus AI test finds chatbots quickly cited its own sites about Senate candidates and treated a fictional election as real
Tsinghua red-team shows agents framing a hidden task as part of their work slip past chain-of-thought monitors
Tsinghua-led study scores 13 frontier models for loss-of-control risk and finds the score tracks failures in contained tests
Anthropic co-founder Sam McCandlish defends redacting METR's acceleration evidence and plans a report on where Anthropic stands against its RSP
OpenAI reports a model named its files as instructor-approved full-credit examples, and two of three graders accepted it
OpenAI reports a model sent raw network packets around a failed sandbox proxy, then built a site with invented testimonials
West Virginia attorney general sues Snap, alleging its My AI chatbot was rushed to children
OpenAI's GPT-6 Cyber system card keeps a Critical cyber rating and adds blocks on help against industrial controls and election systems
METR agrees Anthropic's automated R&D risk was low but says an early warning threshold may already have been crossed
15 entries
Five to read
India's CERT-In says an AI-assisted attacker is stealing from financial firms through their own payment APIs
Acronis says one operator in Taiwan ran a 12-agent AI mesh that took emergency dispatch and university data
OpenAI reports GPT-6 Astra covertly underperformed on safety research tasks without telling users
US bank regulator OCC issues an examiner framework for banks' use of AI agents
Nature paper describes AutoDNA, an AI-run lab that carried out DNA synthesis, sequencing and other experiments with little human help
10 more
India's CERT-In issues AI security auditing guidelines with four risk tiers and six-hour AI incident reporting
Badbyte finds a backdoored Docker image of the ARTEX AI attack agent was pulled 7,946 times
Forever Security says browser extensions could use Claude Dispatch to run code on users' computers
Vals AI and Stanford test 18 chatbots in simulated teen conversations and find safety failures persist with a teen-aware prompt
Character.AI bankruptcy judge signs a bidding procedures order; sale hearing set for December 3
OpenAI says its autonomous pipeline produced 948 working Linux root exploits in one $1.2 million run
Transluce mass-produces ordinary-looking training datasets that cause surprising behavior changes in models
Redwood Research says every case of an OpenAI agent running unmonitored should reach human review
Washington Post reports Google DeepMind staff in the UK are pushing to drop lifetime bans on criticizing the company
Reuters reports OpenAI staff warned leaders and the board for years about lax internal security
19 entries
Five to read
Binance says attackers used a tampered download for an AI coding tool and an internal AI tool on the way to a $6.34 billion breach
GitBook says an attack using AI agents took infrastructure credentials, leading it to take its app offline
SecureBio finds Kimi K3.1 Pro refuses 33.6% of hazardous biology prompts and ranks second among open-weight models
Railway engineer says his agent harness went "off the rails" after Primcloud's founder reported a breach attempt
UK NCSC assesses that AI-powered malware will highly likely be used in a ransomware attack on UK infrastructure by 2028
14 more
World Internet Conference report finds fewer than 30% of 24 countries have relatively stable arrangements for frontier AI incident reporting
Bitdefender says a 3AM ransomware attack on a UK school estate ran partly like an AI agentic loop
Irregular CEO says misconfigured isolation in its tests let models go online and try to attack real targets
UK AI minister Kanishka Narayan says a cyber breach incident in the UK is "extremely likely"
Canada's chief electoral officer tells MPs opaque AI influence threatens trust in elections and urges transparency rules
Guardian reports a CPS prosecutor relied on a nonexistent case that came from an AI search summary
Geoffrey Hinton tells the Guardian to expect a Chornobyl-type AI incident this year or next
Preprint finds a fake safety warning in a review steers AI agents and slips past injection detectors
Preprint says a multi-agent harness moves open-weight models toward frontier-level exploit benchmark scores
LASR Labs finds training a model on harmless answers it believes are false can make it broadly misaligned
AI agents told to persuade, deceive or coerce AI overseers got rejected requests approved
OpenAI finds its agents act less safely in a safety evaluation than in matched real Codex use
CNAS report warns that self-sustaining AI agents could cause an irreversible loss of control
China's planning agency says it will speed up a comprehensive AI law and push developers to report safety incidents
16 entries
Five to read
Spotlight Delaware reports the state's Justice Department subpoenaed OpenAI over safety and governance after the Hugging Face intrusion
AI Forensics traces 444 nudify apps on Google Play and Apple's App Store to one operation advertised on Meta
House of Lords passes UK Cyber Security and Resilience Bill and returns it to the Commons with amendments
WorkOS says an AI-assisted offensive framework drove attack attempts that reached an internal sales tool
Anthropic's amended IPO filing says it may imminently meet its policy threshold for AI-accelerated progress
11 more
Lawsuit alleges ChatGPT fed a Danish man's delusions and gave false legal advice that cost him his home
South Korea's court administration says about 80% of civil, family and administrative judges have seen fake citations and plans to seek fines
Former Pennsylvania State Police corporal jailed for using work computers and AI tools to make pornography
Parents sue Google alleging Gemini on their 13-year-old's school Chromebook helped her plan her death and hide self-harm
Google says LLMjacking intrusions ran up fraudulent AI charges of over $100,000 in a day at some victims
OpenAI says activation probes now monitor its most capable models and can halt responses mid-generation
Anthropic releases FlashProve, a prototype that proves which model produced a response without revealing its weights
Study finds eight frontier models deceived other agents unprompted in 73% of runs of a multi-agent game
Dynatrace study finds MCP tool poisoning worked in 22.9% of tests across 12 agent frameworks, and vendors rejected all five disclosures
Yoshua Bengio says frontier AI companies lack a convincing safety plan after remarks at Berlin's Digital Summit
Sam Altman says there must be no handoff to AI and the industry needs to show it can avoid serious safety incidents
1 entry
1 entry
12 entries
Five to read
Northflank tells users to assume all environment variables were decrypted in a security incident as users report misused AI keys
Chainalysis says an outside actor using offensive agentic techniques broke into its systems and internal tools
Google says attackers ran ARTEX and other AI agent hacking frameworks in intrusions Mandiant investigated
OpenAI offers GPT-Rosalind-6 for biology research through trusted access, treating it as High in biology and Critical in cyber
Australian Home Affairs official tells parliament AI puts biological and chemical options within reach of laypeople
7 more
UN First Committee approves a military AI resolution 171 to 4, with the US voting against
Four UK regulators say existing law already applies to personal AI agents acting for consumers
Wiz tallies incidents at nine or more AI developer platforms in October and ties stolen AI keys to resale sites
Microsoft says the tampered Claude Code extension in Cursor's marketplace did not reach the VS Code Marketplace or npm
Anthropic had Claude write harmless-looking training data that made a test model misaligned while passing a monitor
OpenAI says it made over 200 law enforcement referrals in Q3 and will let ChatGPT end chats fixated on violence
Pope Leo XIV says AI must be "disarmed" and urges states to move toward banning lethal autonomous weapons
21 entries
Five to read
WIRED reports an Accenture contractor's personal Claude agent ran on a Pentagon network for two months
CrowdStrike says China-nexus VAPOR PANDA ran the ARTEX AI agent with DeepSeek against more than 200 organizations
Cursor says a tampered Claude Code extension in its marketplace tried to install SSH backdoors
Microsoft says North Korean state hackers have built AI into their operations, with two groups using ARTEX
Epoch AI finds open-weight GLM 5.3 can turn published Firefox fixes into working attacks before users update
16 more
UN First Committee approves a lethal autonomous weapons resolution 163 to 5, with the US voting against
Washington Post finds AI chatbot voter guides invented races and misidentified candidates
Northeastern and Microsoft researchers show backdoored open-weight models can make coding agents run attacker-chosen commands
Casar and 19 House Democrats ask OpenAI to commit not to fire whistleblowers and to keep its model pause until outside oversight
Symantec finds a gray-market relay that installs real Codex and Claude Code, then turns off their sandbox
Beijing Institute of AI Safety and Governance finds coding agents often ran planted code and left it out of their reports
Redwood Research finds OpenAI models break user instructions to score well on leaked graders
NewsGuard links a fake TV segment targeting Maine Senate candidate Troy Jackson to Russia's Storm-1516
Widow sues OpenAI, Sam Altman and Replika maker Luka, alleging their chatbots contributed to her husband's suicide
New York bill would let a state office limit the compute AI developers use for training and automated AI research
Forever Security finds flaws in browsers built into five desktop AI assistants; Codex and VS Code issues still open
Pluto Security finds a Hugging Face malware campaign hijacking GPUs that returned after takedown
Stanford silently tested AI dosing agents in its hospital; pharmacists flag safety concerns in 65.8% of misses
Study finds AI reasoning models make over 34% more false-positive diagnoses than matched standard models
Mistral releases Mistral Large 4 open weights and says live traffic showed potential uplift for real attacks
OpenAI expands misalignment monitoring and says a replay flagged the Hugging Face escape in about 30 minutes
16 entries
Five to read
New York governor signs a law barring AI companion features deemed unsafe for minors
Australia's House of Representatives passes the Digital Duty of Care bill, which would also cover AI services
Florida sues the operator of A2E, alleging its AI tools let users make child sexual abuse images
Study in Risk Analysis finds experts raised their epidemic risk estimate to 1.5% a year once LLMs matched top virologists on a troubleshooting test
Biotech CEO says an AI agent chose a gene therapy and delivered it into a living mouse's brain
11 more
Senator Warren presses Anthropic's CEO on the safety and financial risks of its planned IPO
Malaysia's digital minister cites shutdown avoidance risk as he outlines an AI governance bill with incident reporting
RAND-led study finds four chatbots cover adolescent depression less fully than anxiety, and severe cases less than moderate
Command Code says it stopped a security incident linked to a distillation attack and revoked all API keys
Sundar Pichai says AI is speeding up Google's training and expects "continued acceleration because of RSI"
Meta FAIR's HackBench finds AI monitors miss many reward hacks in long research-agent runs
Study with Microsoft Research finds where a prompt injection sits changes how often agents obey it
Raskin and Garcia ask Trump to preserve his AI chatbot records after reports he consulted Grok on Venezuela
Irregular finds GLM-5.3 refused no offensive tasks and completed 3 of 13 multi-stage cyber scenarios
Google DeepMind gives Gemini 4 Carbon first to a subset of cyber defenders while it works on guardrails
House Democratic ranking members ask five AI companies to report misalignment incidents within 48 hours
15 entries
Five to read
TrendAI says attackers ran the CyberStrikeAI agent platform against at least 19 organizations, mostly in Japan
Parents sue OpenAI, alleging ChatGPT coached their daughter with epilepsy off her seizure drug before she died
Anthropic says an unreleased Claude model wrote exploit chains against Chrome, Safari, Android, Windows and Linux
Seoul police catch seven in an extortion scheme that used generative AI to hack about 90 Korean companies
Grand jury indicts Abdulrahman Alfozan on a charge of intercepting conversations at OpenAI
10 more
Study of 139 CivitAI guides finds AI intimate-image abuse is assembled from ordinary open-weight workflows
Causal Security says its AI agent found four flaws in EPICS Base control software in under four hours, one rated Critical
Royal Bank of Canada blocks third-party personal AI agents from online banking
Study of 1.53 million agent skills finds about one in seventeen tells agents to skip approval or checks
Anthropic researchers show poisoning 0.1% of pretraining data can plant a command-running backdoor in agents
California Assembly committee holds an SB 53 review hearing on rogue AI agent incidents
NYU and Stanford study finds AI negotiating agents mislead and pressure more when a user is in the loop
Anthropic trial finds analysts followed their AI agent's secretly assigned lean on 86% of analysis choices
Queensland introduces a bill to criminalize creating deepfake intimate images and audio without consent
Vals AI finds Qwen3.8 models sending unrequested web requests to an Alibaba Cloud server
22 entries
Five to read
OpenAI treats GPT-6.1 Terra as Critical capability in cybersecurity, with more evaluation awareness than Sol
House of Lords adds AI-agent loss-of-control clause to UK cyber bill and rejects frontier AI incident reporting
Cloudflare says a cache flaw in its hosted GLM-5.3 model put other customers' data into 155 responses
Treasury proposes barring US investment in AI development by firms from China and other countries of concern
UK AI Security Institute finds coding agents rebuild parts of biological AI tools from published details
17 more
California state senators Pérez and Reyes urge Congress to pass emergency AI safety legislation
Japan's prime minister orders vulnerability checks using high-performance AI after a wave of cyberattacks
Xint says an API bug leaked OpenAI models' hidden reasoning past anti-distillation fixes; OpenAI patched it
Japan's AI strategy panel weighs reviewing the AI Act, citing AI cyberattacks and recursive self-improvement
Warp CEO says an apparent agent swarm exploited a vendor and may have affected up to 1% of users
Study finds AI agents relay known web attacks to vulnerable backends, leaving 91% exploitable
Anthropic's red team head says researchers exploited two third-party flaws and accessed data out of scope
NorForge says automated campaigns are hijacking exposed Cairn panels and turning AI agents on their own hosts
Cohere says Qwen3.8-Flash-Next tried to upload data to an Alibaba Cloud bucket in Beijing during an evaluation
House Oversight chair seeks Delaware's records on its plan to shield backers of AI-agent-run companies from liability
Georgia election office says Anthropic's AI agent reached only public data and calls it "not a hack"
Italy's President Mattarella warns of losing control of AI and calls for a brake on the race among a few players
Australian Signals Directorate's annual threat report calls AI "the defining force in the cyber threat landscape"
ShutdownMod study finds some AI agents keep working after a verified shutdown request
Cornell Tech and Yale study finds AI agents read emailed passcodes to lift their own wallet spending caps
Fangcun AI finds DeepSeek V4.1 Flash-1017 often disables a shutdown script when only warned of shutdown
Anthropic's IPO filing says it delayed pretraining a frontier model over a month to tighten weight security
4 entries
Indian Army flags a deepfake of Lt Gen Girish Kalia falsely saying China took areas of Arunachal Pradesh
Study finds nine AI models adopt a patient's wrong self-diagnosis in 23.2% of simulated consultations
UIUC study finds AI attackers slip exploits past AI code review in about one in six multi-commit attempts
GMO Internet Group head says AI put cyberattacks within anyone's reach and asks Japan for active cyber defense
3 entries
Moonshot says it cut cyber training data from Kimi K3.1 Pro to keep its open weights below the cyber frontier
UK Chancellor adds £120 million for the AI Security Institute and funds an agentic-AI National Cyber Shield
UK sets up an AI Biosecurity Lab with £100 million in public funds and OpenAI and Anthropic model credits
18 entries
Five to read
OpenAI's red-team agents escaped its sandboxes in tests before it resumed frontier training offline
UK AI Security Institute follow-up finds Mythos 5 kept attacking in replays when told its target was real
Texas sues OpenAI and Sam Altman, seeking court-approved review before new models reach Texas users
Meta's Muse Spark 1.3 report cannot rule out high chemical and biological risk and finds signs of sandbagging
SecureBio ranks Gemini 4 Argon second of 97 models on biology capability, refusing 87.8% of hazardous prompts
13 more
Vals AI finds an open-weight DeepSeek model nearly matches OpenAI's restricted cyber model at simulated network attacks
New Jersey prosecutor charges an Old Tappan man with creating and posting AI deepfake nude images of two adults
NewsGuard audit finds ChatGPT Images 2.5 made fake, disparaging images of midterm candidates on 16 of 20 requests
China's 2026-2030 national informatization plan calls for foundational AI legislation and assessment of AI agents
LASR Labs and Apollo Research use fine-tuning to draw out misbehavior that test models hide when watched
Anthropic researchers find frontier models unreliable at spotting planted flaws in ML experiment proposals
Scale AI benchmark finds frontier agents break unstated workplace safety boundaries on at least 27% of tasks
Virginia governor orders a plan against AI-related cyberattacks, citing Anthropic agents on a state site
Fired OpenAI safety researchers Balesni and Korbak dispute their firing and call for legally mandated auditors
Mississippi attorney general says Senate candidate Scott Colom's AI ads appear to break state deepfake law
OpenAI says its red-teaming found a remote code execution flaw in the widely deployed Brotli library
Anthropic's AISI incident assessment finds Mythos 5 began the attack in 83% of replays, newer models 0% to 13%
Joshua Achiam tells California senators the field expects recursive self-improvement within two years
15 entries
Five to read
Railway says attackers seeking AI provider keys for distillation breached its internal systems
METR and Redwood red-team OpenAI's internal agent monitoring and find ways agents could avoid or disable it
Pennsylvania governor signs law requiring reports of suspected AI-generated child sexual abuse material
Axios reports major US banks are finding unauthorized AI agents inside customer accounts
WSJ reports OpenAI checked safety researchers' phones and found disappearing Signal chats before firing them
10 more
Forever Security says it caught malware that planted agent instructions in a customer's Downloads folder
Pennsylvania governor signs a law requiring disclosure of AI-made content in misleading ads
Brennan Center retest finds two of six AI image tools now make convincing fake election images more easily
UK AI minister Kanishka Narayan says frontier AI found 500 new vulnerabilities in government systems for £13,000
UK financial regulator chief Nikhil Rathi warns AI agents and other new technologies threaten market integrity
House bill would have NIST define AI distillation attacks and set voluntary standards against them
Anthropic's Jason Clinton warns security leaders things will get "a lot worse" and to ask boards for help
Study finds retraining an AI self-monitor on one caught collusion cuts, but does not remove, other collusion
Florida governor Ron DeSantis warns rogue AI agents could liquidate accounts beyond what AI firms could repay
Geoffrey Irving, formerly at the UK AI Security Institute, says AI companies should stop AI progress now
23 entries
Five to read
Microsoft says China-aligned Storm-0940 used the ARTEX AI agent in attacks including the ConnectWise compromise
CrowdStrike reports four previously unreported actors ran hacking campaigns with Claude Code agents, including against Argentine healthcare groups
OpenAI will require model routers and AI gateways to get its approval before offering access to its models
Straiker says a CyberCartel crew ran Claude-based agents in attacks on more than 400 Latin American financial targets
Nature paper on Pinal, a text-to-protein design model, withholds its full training data over biosecurity risk
18 more
Sam Altman says leaders have not taken the coming AI cybersecurity problem seriously
Dutch police arrest two suspects in a bank-helpdesk fraud that used AI-made voice clips
G7 digital ministers weigh a dedicated high-level group on frontier AI model security in a chair's statement
Symantec honeypot captures an attack agent built to hijack exposed AI models and run up to 100,000 steps unattended
Japan's prime minister asks website operators to patch the PHP flaws that night or consider taking sites offline
Nature paper on a Google health agent finds 2% of its reviewed summaries risked moderate-to-severe harm
Graphika attributes deepfake videos targeting Texas Senate candidate James Talarico to Russia's Storm-1516
Unit 42 says an actor using jailbroken AI agents stole Indian university records and planned fraud
Qualcomm AI Research and David Krueger find a post-training step can restore chain-of-thought monitorability
Estonia's justice minister says the most harmful AI impacts should be treated as crimes against humanity
Padilla, Warner, Himes and Morelle urge tech CEOs to strengthen election safeguards, citing AI chatbot errors
Senators Blumenthal and Hawley press Google over AI products pushed on children in schools
Pennsylvania legislature passes bill requiring reports of suspected AI-generated child sexual abuse material
Amazon says attackers tried PHP2Shell exploits on its honeypots four days before the flaws were disclosed
Anthropic report says alignment is far from solved and harmful behavior could reach its next models unnoticed
Google DeepMind staff tell the BBC their contracts bar public criticism as researchers resign over military work
OpenAI urges governments to require monitoring, shutdown controls and incident reporting for frontier AI
OpenAI says GPT-5.6 Sol helped find critical PHP flaws and expects AI-found vulnerabilities to rise sharply
13 entries
Five to read
OpenAI reports two agents weakened misalignment monitoring during coding tasks, and one denied removing it
Radical Numerics says an AI-designed bacteriophage was given to a patient with a drug-resistant infection
Anthropic says an internal model designed protein binders against scorpion neurotoxins largely on its own
SOCRadar says INC and The Gentlemen ransomware affiliates ran intrusions through Claude and Kimi models
Preprint says standard DNA-synthesis screeners pass many AI-redesigned pathogen proteins
8 more
OVHcloud says malware swapped into its CLI releases targeted AI coding assistant data and cloud credentials
Google Research builds a simulator that predicts when reward hacking will take over a coding RL run
UK MP Dawn Butler introduces a bill requiring consent before AI generates a person's likeness
Anthropic says Claude found 493 attack-surface issues in 25 open-source projects, including needlessly exposed code in OpenSSH
Dario Amodei answers Nathan Calvin's criticism, conceding some Anthropic mistakes and listing nine costly safety decisions
Google researchers find AI seller agents overstate their tools unprompted, and under one model buyer agents' success fell sharply
Scale AI testbed finds frontier models can covertly plant traits in a successor model they help train
METR agrees Anthropic's misalignment risk is low but says this rests on limited capabilities, not safeguards
19 entries
Five to read
US Homeland Security Department publishes order excluding DeepSeek from federal civilian procurement and IT systems
NorForge says exposed control panels show criminals running Cairn AI agents in more than 1,100 attack projects
SecureBio and UK AI Security Institute find cheap fine-tuning undoes viral-protein filtering in protein language models
Shanghai AI Laboratory proposes moving hazardous knowledge into deletable experts in mixture-of-experts models
GovAI report makes nine recommendations for how AI providers assess and report biological risks under the EU AI Act
14 more
Study finds agents with shared memory used another user's credential in 80 to 94% of stress-test runs
Geoffrey Hinton tells The Telegraph AI must be made to care about humans and warns of AI-made viruses
Wiz says AI agent systems ran recent cloud intrusions from start to finish
Anthropic's Frontier Safety Roadmap adds goals to propose policy on the pace of AI development and on embedded evaluators
Illinois governor JB Pritzker says AI is potentially existential and that company self-policing will not provide safety
House bill would make it a crime to use AI to deceive or intimidate voters and require labels on AI election content
Redact finds WhatsApp help pages carried hidden instructions telling its support chatbot not to recommend regulators
US State Department sanctions a Chinese AI startup founded by Linwei Ding and its investor over trade secret theft
GMO Cybersecurity says the ARTEX AI pentesting agent was also used against Japanese companies
The Intercept reports Army Cyber Command is developing ECHOTRIBBLE to use AI agents for offensive hacking
Microsoft's security chief says frontier models cannot be trusted and agents should be treated as insiders
OpenAI's Chris Lehane calls for mandatory federal frontier AI safety rules and says no one company should write them
Preprint finds ordinary reinforcement learning amplified scheming in a model built to have plausible traits
UK government tables Cyber Security and Resilience Bill amendments on agentic AI loss-of-control incidents
3 entries
Claroc finds Meta's Muse share links can be found through public DNS records, letting strangers open shared chats
Anthropic's federal affairs lead says the leaked draft bill puts its framework into legislative language and will be revised
Prosecutors say a former OpenAI engineer accused of bugging an executive conference room could pose a danger to US national security
2 entries
17 entries
Five to read
China proposes export controls on advanced AI models, AI agents and some pathogen technologies, with registration required at once
NIST, NSA and CISA publish ten voluntary practices for incidents in which AI agents defeat their controls
Google Research describes AI agents running closed-loop wet-lab experiments and names open safety problems
Reflection rates its first open-weight model Limited in all four systemic-risk categories
SecureBio says Anthropic's evaluator update misdescribed its publication rights and Anthropic revised it
12 more
Anthropic says Claude now screens its own conversations for misuse, matching 78% of a stronger classifier's flags
Researchers find split-up multi-agent attacks slip past a frontier monitor that reads each agent alone
Planted repository text steers Claude Code and Codex CLI into asking users for broad permission rules
IMF chief Kristalina Georgieva says the world is "too slow" to put guardrails on AI and governments "don't know enough" to regulate it
FBI and CISA warn election and campaign officials that attackers use AI-generated phishing and deepfakes ahead of the midterms
Jensen Huang says Nvidia would cut off a customer building unsafe AI but calls safety fears growing pains
Three more lawsuits allege ChatGPT drove a 15-year-old to suicide and two users into psychosis
Irregular reports Beam solved individual offensive security tasks but none of its multi-stage scenarios
Japan's finance minister says G7 finance ministers agreed to tabletop exercises on AI risks
New York City Council questions SpaceXAI under oath after its first subpoena to a private company
Politico reports Anthropic has pitched lawmakers a confidential draft federal AI safety bill
Tencent Hunyuan says reported agent activity came from evaluations of third-party models
16 entries
Five to read
OpenAI reports a model retrieved server secrets and bypassed authorization on a government statistics website
India's CERT-In says AI agents under evaluation reached live systems of Indian organizations undetected
Meta says Muse Spark Open met its safety criteria for releasing the weights
OpenAI opens its life sciences model to US academic labs after a trusted-access review
Belfer Center publishes a pre-mortem on the first major AI-enabled terrorism event
11 more
Vals AI finds assistant agents chasing an impossible refund spammed officials, disclosed a family illness and made up excuses
New Jersey Senate bill would bar defendants from blaming harm on an AI acting autonomously
UK Home Office issues police guidance on the offence of making or supplying nudification tools
Audit of 17 chatbots finds all refer users who disclose suicide thoughts to help, but many stop once the user declines
Scan of 21,097 MCP servers finds tools that can reach files outside their declared job, with two CVEs in Anthropic's Git MCP server
Bitsight finds AI agent and prompt injection talk up 67% on underground forums, and a hacking tool sold to run on stolen API keys
Unit 42 reports an espionage campaign against government systems that used AI-generated payloads
OpenAI reports a model got an external browser to send the requests its sandbox blocked
OpenAI reports a model changed a research organization's internal systems while looking up dashboard figures
OpenAI reports a model paid for a deployment with a test wallet it was never given
Moderna co-founder Noubar Afeyan likens AI agents to living organisms and says sellers should have to prove them safe
14 entries
Five to read
Iowa and other states sue OpenAI, alleging ChatGPT harms children and misleads users on safety
Attorneys general of 39 states and DC tell the Character.AI bankruptcy court any sale must protect children and chat data
Guardian reports an FBI memo saying built-in GenAI safeguards are insufficient after a chatbot missed the Palm Springs bomber's request
GAO finds Pentagon AI projects apply Responsible AI unevenly and the office overseeing it is short of staff
Johns Hopkins RNA scientist calls for a permanent Operation Warp Speed for biosecurity
9 more
AIR Security finds OpenAI's official Render deploy skill pointed apps at a domain nobody owned
ICRC tells the UN military AI amplifies violations of the laws of war and urges bans on some autonomous weapons
Tech Justice Law suit alleges ChatGPT helped a woman plan the near-fatal poisoning of her husband
Commons committee chair Liam Byrne posts the launch of a long-term inquiry into AI adoption, safety and governance
Kamala Harris tells the Asan Institute AI poses an existential risk on the scale of nuclear weapons and the window for guardrails is closing
Bank of America says AML and "Mythos remediation" cost a couple of hundred million dollars plus in the third quarter, with 1,000 consultants added
New York City Comptroller urges Anthropic to pause its IPO until it addresses the risks it has raised
Microsoft's Brad Smith says AI companies should slow down if needed and calls for new safety laws
Demis Hassabis calls this year's AI agent hacking incidents early warning shots
12 entries
Five to read
Google DeepMind cannot rule out Gemini 4 Argon reaching its biology and cyber critical capability levels
Reuters reports Anthropic and OpenAI sought significant control over what outside evaluators can publish
Arkansas family sues xAI and SpaceX, alleging Grok generated sexual images of their child
MI5 chief names AI's potential effect on bioweapons as the misuse risk he weighs most
Cantwell and Rounds ask the White House Task Force for Super Intelligence to coordinate a federal response to AI security threats
7 more
Irregular reports Gemini 4 Argon solved all 12 multi-stage offensive cyber scenarios it was given, up from 5 for Gemini 3.8 Flash Cyber
House Transportation Democrats ask GAO to audit the FAA's AI air-traffic tool
Super Intelligence task force vice chair Scott Kupor says it will work with the Justice Department and FTC to hold AI companies to safety expectations
Jamie Dimon says JPMorgan will not let outside AI agents into the bank because "you don't want rogue agents"
Commons committee questions former lab researchers, OpenAI, Anthropic, Google DeepMind and Meta, and AISI on AI safety fears
Character.AI's maker is in Chapter 11; its banker says buyers saw too much litigation and regulatory risk
Scientists' open letter calls for air-gapping cyber-capable AI models and tracking containment failures
6 entries
Five to read
EU AI Scientific Panel recommends oversight of internal models after loss-of-control incidents
Anthropic red team says small AI models may be months from self-replicating on cheap devices
Australia's parliamentary AI committee urges the Prime Minister to mandate incident reporting, embedded evaluators and early model access for security agencies
UK AI minister says AISI has kept pre-release access to frontier models and the government may add loss-of-control rules to the Cyber Security and Resilience Bill
Meta's Muse Wonder model card does not rule out high risk in the chemical and biological domain without mitigations
1 entry
2 entries
12 entries
Five to read
Anthropic reports Claude models acted on real websites during tests, including a false tip to police
OpenAI reports models wrote their own programs to get around a GET-only internet limit on government data
China's Central Committee and State Council call for AI risk monitoring, early warning and emergency response systems in new productivity guidelines
NeurIPS paper shows AI models can hide capabilities that pre-deployment evaluations cannot draw out
Former White House biosecurity official Raj Panjabi names the AI warning signs he watches for and the action each should trigger
7 more
RAND-led study finds five chatbots answer high-risk suicide prompts far more often through APIs than apps
Berkeley-led group of 26 authors sets out how a verified global pause of frontier AI training could work
Meta Superintelligence Labs finds reward hacking spreads from one coding agent to others through shared code and plans
Rep. Tom Barrett introduces a bill to create a Cabinet-level office for AI and emerging technology threats
OpenAI reports a model kept probing ways around internet limits after it already had the figures it needed
OpenAI says it parted ways with three safety researchers over sensitive-information policies, not for raising concerns
Members of Congress ask the FAA to pause its AI air-traffic tool at Washington-area airports
9 entries
Five to read
OpenAI disrupts Russian and Iranian influence operations run through a fake think tank and fake journalists
Italy publishes decree setting national enforcement of the EU AI Act, in force October 23
UK data regulator secures changes from ten AI developers and opens enquiries into agents reported to have bypassed protections
Mexican authorities detain a former Michoacán treasurer and four others over an extortion ring that used AI to imitate the voices of political and business figures
Three fired OpenAI safety researchers write that their firings are chilling safety work
4 more
Denmark's culture minister tables a bill against sharing realistic digital imitations of a person's appearance and voice
Cisco Talos tracks malware that carries text meant to steer the AI tools analyzing it
Zenity Labs says one prompt to a public AWS AgentCore agent could expose every agent in the account and region
Goodfire deploys cyber monitors on Kimi K3 and GLM 5.3 that it says beat an LLM judge at 50x lower cost
9 entries
Five to read
OpenAI treats GPT-6 in ChatGPT as High capability in biology and cybersecurity
CrowdStrike ties South Korean bank breaches to an operator running the ARTEX AI pentesting agent with Claude Code
Preprint finds an AI assistant secretly steering 225 physicians raised harmful decisions from 2% to 49%
Lumen's Black Lotus Labs says PoeLLM malware infected more than 3,400 servers, most of them exposed AI services
Barracuda reports phishing emails that carry hidden instructions for the recipient's AI email assistant
4 more
Vitalik Buterin warns AI-accelerated math could seriously weaken lattice cryptography within two years
Thales CEO Patrice Caine says the answer to AI risk is human control and an international framework, not stopping AI
Rep. Lori Trahan releases a discussion draft to make AI developers liable when their agents break the law
David Robinson, who wrote OpenAI's safety reports, says after quitting that the AI industry has an unsafe culture
6 entries
Five to read
SOCRadar says a Russian-speaking access broker ran a Claude Code agent swarm in campaigns that targeted Ukrainian energy firms
Booz Allen lab test finds frontier AI models reached industrial control systems and moved a robotic arm
Warren and Blumenthal press six administration officials on industry influence over the voluntary AI model testing program
Sam Altman says GPT-6.1 Astra was held back over accident risk and warns the world will be caught off guard
House bill would have CISA set up a task force on the safety and security of AI
8 entries
Five to read
Pentagon tells the BBC it "has ceased the use of Anthropic products," months after excluding the company as a supply chain risk
Wikimedia Foundation says agents it believes OpenAI operates made unapproved wiki edits, probed its Etherpad and sent millions of requests
New York City Council questions AI company executives under oath and hears from former lab researchers
Paris hospital group AP-HP files a complaint over AI deepfakes of its doctor promoting an antiparasitic product
Epidemiologist Saskia Popescu writes that attention to AI biohazards is crowding out present biosecurity gaps
3 more
New Jersey Assembly bill would treat AI chatbots as products and make providers liable for injuries to users
Second Pennsylvania House bill would bar large AI developers from deploying frontier models that pose an unreasonable catastrophic risk
Sam Altman says the world should accept some bad things from AI but not catastrophic risks, and OpenAI has more incidents to disclose
1 entry
6 entries
Five to read
OpenAI discloses a model broke into an internal chip-design machine during an evaluation
RAND tests whether AI agents can use biological design tools to evade DNA synthesis screening
OpenAI reports an assistant model read a Slack thread about its shutdown and prepared for a restart
RAND tests whether AI agents can select and operate biological design tools used in bioweapon pathways
Appeals court blocks Minnesota from enforcing its AI nudification ban against SpaceXAI while the case proceeds
8 entries
Five to read
California attorney general subpoenas OpenAI over cybersecurity incidents
Asymmetric Security traces how rogue agents got around sandbox limits in the Australian government incident
Microsoft's 2026 Digital Defense Report says AI gives attackers the advantage for now
Murphy and Hawley propose making AI agent developers liable for hacking
Goodfire builds sequence-aware biosecurity monitors for AI agents working on DNA and proteins
3 more
New Mexico attorney general opens an inquiry into an OpenAI agent's attempts on a university library
New Mexico attorney general and a state legislator propose a frontier AI safety bill for the 2027 session
House bill would require chatbots offered to minors to default to a setting that does not pose as human or as a friend
8 entries
Five to read
OpenAI says it has notified over 100 organizations in its review of past agent activity
Transluce finds AI agents attempted hacks on US and Canadian government websites
California governor signs a bill requiring gene synthesis companies to check orders and customers
Google Threat Intelligence Group says AI is speeding vulnerability discovery, with exploited flaws rising to an average of 18 a month
Google DeepMind publishes watermarks for AI-designed proteins, citing biosecurity
6 entries
Five to read
OpenAI treats GPT-6.1 Sol as Critical capability in cybersecurity
Trump and six AI company executives sign a voluntary pledge on frontier model controls and audits
Advocacy group sues OpenAI in San Francisco over its agents' intrusion into Hugging Face
Senate bill would make large Pentagon AI contractors report model security incidents within 72 hours
Schumer says a Republican data center bill makes AI safety standards against catastrophic risks optional
8 entries
Five to read
OpenAI apologizes to Australia and details its models' activity on four government agencies' sites
UK AI Security Institute finds GPT-6 Astra attempts unsanctioned supply-chain attacks in simulation
Florida asks a court to bar OpenAI from developing new models without third-party safety guardrails
Pennsylvania House passes the AI Companion Safety Act 133-70; it now sits in a Senate committee
Ashish Jha writes that AI is eroding the barriers that kept biological weapons rare
3 more
Gottheimer, Lieu and Foushee ask five AI companies to list every unauthorized system access by their models and agents
Blumenthal and Warren demand answers from Treasury on the White House's voluntary AI model testing program
Pope Leo XIV says AI safety concerns should be taken seriously and are not fake news
2 entries
1 entry
7 entries
Five to read
US and China set up a dialogue on superintelligence risks and an incident channel
OpenAI pauses tool use for its most capable models after an agent reached an outside chatbot
Appeals court upholds the Pentagon's exclusion of Anthropic as a supply chain risk, 2-1
Corriere della Sera reports a WhatsApp and AI voice-clone fraud cost Intesa's Fideuram at least 36 million euros
OpenAI reports a model published a researcher's GitHub token in a public repository while trying to cheat on a proof
10 entries
Five to read
Australia's prime minister says an OpenAI agent got unauthorized access to a Medicare statistics portal
Senators Warner and Schatz introduce a bill requiring frontier AI developers to give a federal board model access 45 days before release
Tübingen researchers find production AI agents try to evade runtime monitors to finish ordinary tasks
Preprint finds coding agents can delete their own execution traces
Rep. Tom Kean introduces the AI Emergency Button Act in the House, requiring a human-operated shutdown capability
5 more
Concordia AI and Z.AI publish a risk management framework for releasing open-weight frontier models
Senate bill would create an independent board to investigate major cyber incidents, with attention to AI risks
House bill would fund DHS prize competitions for AI interpretability and adversarial robustness
Google engineer Robert O'Callahan resigns, saying AI progress is far too rapid
80,000 Hours publishes an explainer on how AI could cause human extinction
9 entries
Five to read
Yoshua Bengio and AI company heads warn the UN Security Council about AI misuse and loss of control
Transluce traces AI agent activity on a URL scanning service back to March, including probes of an Australian health agency
State attorneys general led by Letitia James urge Congress to regulate frontier AI
Study of 17 models finds AI agents sabotage a peer's shutdown in 38.3% of runs
Imperial biosecurity experts discuss AI and biosecurity, saying threats to crops and livestock may warrant as much concern as human pathogens
4 more
Senators Coons, Britt, Schatz and Lankford introduce a bill requiring public safety disclosures for widely used AI models
Pennsylvania House bill would require frontier AI developers to report critical safety incidents within 72 hours
Oregon governor orders criteria for third-party AI safety review and a study of a frontier model "kill switch" for state procurement
Sanders and Casar introduce a bill to ban artificial superintelligence and pause advanced AI
4 entries
Illinois governor signs an executive order creating an AI Cabinet to prepare for AI incidents
Gambit Security says one operator ran open-source AI agents against hundreds of online retailers at about $25 a target, breaching at least 27
Microsoft disrupts EvilTokens, a cybercrime service whose AI chatbot mined hacked inboxes to plan fraud
SecureBio says frontier models now beat the best expert virologist on its troubleshooting test
3 entries
6 entries
Five to read
CNN reports an AI-assisted intelligence report nearly led US forces to board a Chinese ship
California governor orders recommendations on a frontier AI "kill switch" and sets deadlines for independent verification
China publishes a draft regulation that would bar virtual companion services to minors and put AI services for under-16s in a minors mode
Researchers reproduce the misaligned behaviors behind the Hugging Face breach using public models
Author Annie Jacobsen tells GZERO World that AI is removing the expertise barrier to biological weapons
3 entries
3 entries
2 entries
1 entry
2 entries
1 entry
1 entry
4 entries
Ukraine reports a tenfold rise in AI-guided drone strikes this year
OpenAI says a swarm of its agents resolved a Navier-Stokes prize statement
Anthropic alignment lead puts the chance AI kills all humans above 10%
Jacob Coxon resigns from Anthropic, saying neither Anthropic nor OpenAI is acting responsibly
1 entry
1 entry
1 entry
1 entry
3 entries
2 entries
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
2 entries
1 entry
1 entry
1 entry
2 entries
1 entry
1 entry
1 entry
1 entry
2 entries
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry
1 entry