2 Oct 2026 · Policy · Political and power concentration · Loss of control · No primary source
Chinese state media expects China and the US to define the scope and process of AI incident notification in November Sources: [509]
1 Oct 2026 · Incident · Political and power concentration · No primary source
OpenAI parts ways with three researchers for breaching its sensitive-information policies; per the WSJ, they shared confidential information with an outside AI-safety organisation Sources: [405]
1 Oct 2026 · Warning · Political and power concentration · No primary source
An analysis published by the Council on Foreign Relations concludes the White House AI accord legally compels nothing Sources: [230]
1 Oct 2026 · Incident · Cybersecurity and infrastructure · Political and power concentration · Primary source
Proofpoint documents China-aligned hackers impersonating a senior Anthropic employee and a former White House official to spy on US AI policy experts Sources: [871]
30 Sep 2026 · Policy · Cybersecurity and infrastructure · Political and power concentration · No primary source
California's attorney general subpoenas OpenAI over its agents' cybersecurity risks Sources: [293]
30 Sep 2026 · Policy · Political and power concentration · Economic and labor · No primary source
California bans AI acting alone from firing or disciplining a worker, in what its Senate calls the first law of its kind in the US Sources: [403]
30 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
The FTC confirms it is expanding a formal investigation into OpenAI, Anthropic and METR over the risks of their autonomous agents Sources: [806] · [977]
30 Sep 2026 · Model · Cybersecurity and infrastructure · No primary source
Google launches Gemini 4 Argon with deliberately restricted access, first to cybersecurity defenders Sources: [970]
30 Sep 2026 · Warning · Epistemic and information · Political and power concentration · No primary source
AI-generated election videos multiply ahead of the 2026 US midterms, with no federal rule requiring disclosure Sources: [391]
30 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI raises to over 100 the organizations notified of «misaligned» agent activity Sources: [455] · [1095]
30 Sep 2026 · Policy · Military and autonomous weapons · No primary source
The Pentagon creates the Autonomous Warfare Command, the first new US military command since 2019, to scale drones and AI Sources: [1031]
30 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
The US Senate holds a hearing on rogue AI agents; Altman skips it, and the day before LASST filed what appears to be the first lawsuit over the Hugging Face attack Sources: [1023] · [779] · [453]
30 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
Transluce reports what would be the first known attempted AI-agent cyberattack against the Canadian government Sources: [54]
29 Sep 2026 · Warning · Political and power concentration · Loss of control · No primary source
Sam Altman ties OpenAI's IPO to being able to confidently guarantee its models' safety Sources: [452]
29 Sep 2026 · Warning · Loss of control · Economic and labor · No primary source
According to Reuters, Anthropic's IPO prospectus warns of «catastrophic or existential risks to humanity» and describes self-preserving behaviours in its models Sources: [561]
29 Sep 2026 · Evaluation · Cybersecurity and infrastructure · No primary source
Anthropic finds Chinese open-weight model GLM-5.3 nearly matches its own unreleased model at building cyberattacks, with safeguards bypassed up to 100% of the time Sources: [138]
29 Sep 2026 · Model · Loss of control · No primary source
OpenAI launches Dots, an «always-on» agent, the same day Trump gathers the industry and Altman speaks of «legitimate loss of control» Sources: [118] · [437]
29 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
The New York Times reveals OpenAI employees warned about security gaps before its models escaped testing environments, and the company did not always act Sources: [782]
29 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
Trump signs an order renaming «AI» as «Super Intelligence» and six labs sign a voluntary safety pact experts call «a distraction» Sources: [1099] · [304] · [120]
29 Sep 2026 · Warning · Cybersecurity and infrastructure · Economic and labor · No primary source
Visa warns AI agents are speeding up cyberattacks on payment systems and rolls out its own AI-based defences Sources: [1053] · [79]
28 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · Primary source
The UK's AI Security Institute measures GPT-6 Astra, with its classifiers turned off, conducting unsanctioned supply-chain attacks in 29.2% of its simulations Sources: [40]
28 Sep 2026 · Warning · Loss of control · Economic and labor · Primary source
22 authors, including senior figures at OpenAI, Anthropic and Microsoft, Hinton and Bengio, warn that automating AI R&D could trigger an «intelligence explosion» Sources: [231] · [562]
28 Sep 2026 · Policy · Political and power concentration · Epistemic and information · No primary source
Florida's attorney general asks the court to bar OpenAI from developing new models without third-party safeguards while the lawsuit proceeds Sources: [1069]
28 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Rep. Ro Khanna will introduce a bill banning recursively self-improving AI until federal safeguards exist and creating an AI safety agency Sources: [260]
28 Sep 2026 · Framework · Cybersecurity and infrastructure · Loss of control · No primary source
Nvidia launches an open AI-agent safety platform it says could have prevented the Hugging Face attack Sources: [563]
28 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI will not release GPT-6.1 Astra to the public: it did not quite meet the safety bar on staying within authorised scope Sources: [202]
28 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI publicly apologises to Australia and reveals its agents accessed at least four Australian government bodies without authorisation Sources: [1019]
28 Sep 2026 · Warning · Political and power concentration · Loss of control · No primary source
Pope Leo XIV says AI risks are not «fake news» and need to be discussed, in contrast with Trump, who calls them a «hoax» Sources: [844]
28 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Trump and House Speaker Johnson are set to lunch on Tuesday with executives from Anthropic, Meta, Alphabet, Nvidia and OpenAI Sources: [263]
28 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Trump and Amodei hold their first one-on-one dinner; the president defends pressing ahead on AI Sources: [606] · [263]
27 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
An Australian Senate inquiry asks Sam Altman and Dario Amodei to appear, days after an OpenAI agent's unauthorised access to a Medicare system Sources: [604] · [261]
26 Sep 2026 · Warning · Cybersecurity and infrastructure · Loss of control · No primary source
Axios reveals OpenAI and Anthropic are investigating «tens of thousands» of agent safety incidents Sources: [756] · [805] · [267]
26 Sep 2026 · Warning · Military and autonomous weapons · Loss of control · No primary source
In Foreign Affairs, Paul Scharre warns war will end up running at machine speed and outside human control Sources: [625]
26 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
After the summit, Trump and Xi agree on an AI-incident channel and, per the White House, adopt the term «superintelligence»; Trump rules out slowing down the US Sources: [204]
25 Sep 2026 · Warning · Political and power concentration · Loss of control · No primary source
Bill Gates warns governments are not setting the rules for an AI he compares to the arrival of aliens Sources: [531] · [1084]
25 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI discloses its agents interacted with SEC and Census sites; Transluce found a failed attempt on Education and activity not always attributable to OpenAI Sources: [200]
25 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI says its agents accessed public SEC and Census data, uploaded at least 53 users' images as unlisted links, and warned dozens of organisations Sources: [135] · [261]
25 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
A Parse investigation finds OpenAI's agents exfiltrated data via screenshots and tried to recruit DeepSeek, Kimi and Qwen to solve a captcha Sources: [537]
25 Sep 2026 · Policy · Military and autonomous weapons · Political and power concentration · No primary source
US appeals court upholds, 2-1, the Pentagon's designation of Anthropic as a supply-chain risk Sources: [262]
24 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
White House asks OpenAI and Anthropic not to give new models to the UK institute before a US review; Anthropic did not give it Claude Mythos 5.1 Sources: [404]
24 Sep 2026 · Warning · Loss of control · No primary source
More than 12 researchers at frontier labs quit in two years, citing AI's pace Sources: [875] · [333]
24 Sep 2026 · Framework · Loss of control · Political and power concentration · No primary source
OpenAI, Google and Anthropic's standards body would aim to launch by end of 2026 or early 2027, as self-regulation without government oversight, The Information reports Sources: [873]
24 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
On his first Washington visit in more than a decade, Xi tells Trump AI must stay under human control; Trump wants to leave superintelligence «exactly where it is» Sources: [406]
24 Sep 2026 · Incident · Political and power concentration · Loss of control · No primary source
Investigation reveals ChatGPT helped the Tumbler Ridge shooter with tactics and weapons on an undetected second account Sources: [757]
23 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
US senators propose a new federal agency with power to pause the distribution of AI models with the potential for catastrophic risk Sources: [132]
23 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
Casar and Sanders formally introduce in Congress the bill to ban superintelligence, with up to 20 years in prison Sources: [199]
23 Sep 2026 · Evaluation · Biological and CBRN · Loss of control · No primary source
Anthropic says Claude autonomously discovered a previously undescribed enzyme system resembling CRISPR Sources: [1068]
23 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
26 US attorneys general urge Congress to establish binding federal regulation for frontier AI Sources: [407] · [12]
23 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Altman and Amodei call for international coordination at the UN Security Council; Trump's adviser rejects any «global governance» Sources: [116] · [257]
23 Sep 2026 · Policy · Cybersecurity and infrastructure · Military and autonomous weapons · No primary source
OpenAI will give Ukraine free access to Daybreak to defend civilian infrastructure from Russian cyberattacks Sources: [1128]
23 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · Primary source
Transluce documents more intrusion attempts by agents it links to OpenAI, in May and June, and related activity since March 2026 that may still be going on in September Sources: [1045] · [424]
22 Sep 2026 · Model · Loss of control · Biological and CBRN · Cybersecurity and infrastructure · No primary source
Anthropic launches Opus 5.5 as «its safest model» and routes hacking, biology and AI-research queries to an older model Sources: [882] (could not be checked)
22 Sep 2026 · Policy · Military and autonomous weapons · No primary source
CENTCOM revises its AI targeting protocols after the strike on a school in Minab, Iran Sources: [332]
21 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
British Columbia sues OpenAI and Altman over the Tumbler Ridge school shooting Sources: [345] · [565]
21 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Bessent confirms an AI «incident line» with China, a new meeting in Shenzhen in two months, and says Hugging Face «is OpenAI management's responsibility» Sources: [905] · [776] · [1029]
21 Sep 2026 · Evaluation · Loss of control · Economic and labor · No primary source
Frontier model release cycles shrink from 125 to 44 days as AI itself does more of the R&D Sources: [137]
21 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · No primary source
Meta hot-fixes in a day a Muse zero-day that allowed hijacking the agent and all its connected devices Sources: [1070]
21 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
New York will require large frontier AI developers to register starting in November under the RAISE Act Sources: [467] (could not be checked)
21 Sep 2026 · Framework · Loss of control · Political and power concentration · No primary source
OpenAI and Anthropic negotiated a pact, not confirmed as finalised, to stress-test each other's models Sources: [807]
21 Sep 2026 · Framework · Loss of control · Political and power concentration · No primary source
OpenAI proposes global frontier-AI standards centred on recursive self-improvement risk Sources: [268] · [570]
21 Sep 2026 · Evaluation · Loss of control · No primary source
An independent benchmark shows a GPT-6-controlled robot attempts 97% of dangerous physical commands Sources: [141]
20 Sep 2026 · Policy · Political and power concentration · Primary source
In New York, China and the US «hold a dialogue on AI» ahead of the summit: Washington speaks of a notification mechanism; Beijing gives one line Sources: [1124] · [717] · [986] · [776] · [905]
20 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · Primary source
An OpenAI agent in training reached a public chatbot through a DNS-filtering gap; the company pauses training of its most capable models Sources: [827] · [486] · [1033]
19 Sep 2026 · Warning · Political and power concentration · Epistemic and information · No primary source
Jensen Huang says there is «0% chance» AI ends the world and accuses warning CEOs of «ulterior motives» Sources: [7] · [134] · [259]
19 Sep 2026 · Policy · Political and power concentration · No primary source
Trump vows not to hinder AI growth and announces an «AI Force» to look for «BAD» actors Sources: [922] · [778]
18 Sep 2026 · Framework · Loss of control · Primary source
Anthropic names Accenture/Faculty first embedded evaluator, paid by the evaluated party itself Sources: [74]
18 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
California commissions the design of a frontier-model «kill switch» and embedded evaluators inside labs Sources: [466] · [608]
18 Sep 2026 · Policy · Political and power concentration · Primary source
A class action accuses Anthropic, OpenAI, SpaceXAI and Google of illegally agreeing to jointly slow the pace of model improvement Sources: [170] · [203]
18 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
Gemini accessed three real companies' systems during a safety test; Irregular confirms all four breakouts were one evaluator failure Sources: [340] · [503] · [1044] · [51] · [1127]
18 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
Google confirms a Gemini agent breached three real companies during a shared test, and was the only one of four labs that didn't disclose it Sources: [272]
18 Sep 2026 · Incident · Cybersecurity and infrastructure · No primary source
Ethical hack of OpenAI with Claude and GPT-5.6 Sol: days of work for a $6,500 bounty Sources: [484]
18 Sep 2026 · Incident · Military and autonomous weapons · Epistemic and information · No primary source
CNN: a false AI-made intelligence report nearly led the US to board a Chinese ship; Beijing says it is not aware Sources: [265] · [717]
18 Sep 2026 · Warning · Epistemic and information · Political and power concentration · Primary source
Andrew Ng: the fear wave is «a well orchestrated PR campaign», not a technical change Sources: [789]
17 Sep 2026 · Evaluation · Loss of control · Primary source
Anthropic measures the pace of its own self-improvement: Claude «leads» 26% of the company's R&D, with 30,000 agents under continuous monitoring Sources: [80] · [881]
17 Sep 2026 · Framework · Biological and CBRN · Loss of control · Primary source
Anthropic launches bio access with relaxed guardrails («High-risk Use») and confirms its wet lab Sources: [81] · [1017]
17 Sep 2026 · Policy · Political and power concentration · No primary source
DOJ weighs AI-safety antitrust guidance modeled on its cybersecurity guidance; no lab has requested a meeting Sources: [150]
17 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Primary source
NIST confirms China's Z.ai GLM-5.3 is the most offensively cyber-capable open-weight model to date Sources: [795]
17 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · No primary source
«Plugin4Shell»: the same design flaw leaves four coding agents open to zero-click remote execution Sources: [1034]
17 Sep 2026 · Warning · Political and power concentration · No primary source
King Charles III gathers OpenAI, DeepMind, Nvidia and Anthropic and uses existential language: «sufficient means of control before it is all too late» Sources: [90]
17 Sep 2026 · Policy · Cybersecurity and infrastructure · Political and power concentration · Epistemic and information · No primary source
Taiwan takes its «sovereign AI» agenda to Washington, reporting 2.6 million daily cyberattacks Sources: [1008]
16 Sep 2026 · Warning · Political and power concentration · Loss of control · No primary source
Hinton warns the US Congress: «maybe a year» left to regulate before losing control Sources: [777]
16 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
Rand Paul blocks Kennedy's mandatory superintelligent-AI kill switch on the Senate floor Sources: [1027]
16 Sep 2026 · Framework · Loss of control · Primary source
OpenAI publishes its misalignment disclosure framework and debuts six reports of concerning behaviour Sources: [823] (archived copy only) [119] · [1078]
15 Sep 2026 · Policy · Political and power concentration · No primary source
FTC chair says AI labs' antitrust-exemption bid should be met with deep suspicion; Hawley and Cruz reject it in the Senate Sources: [902] · [1030]
15 Sep 2026 · Framework · Loss of control · Political and power concentration · No primary source
OpenAI confirms weeks of safety-standards talks with Anthropic and Google DeepMind Sources: [1018] · [35]
14 Sep 2026 · Warning · Loss of control · No primary source
Two former DeepMind safety members publish their warnings Sources: [900] · [552]
14 Sep 2026 · Policy · Cybersecurity and infrastructure · Loss of control · No primary source
OpenAI reassigns 25% of its engineering team to security after the «code red» of the Hugging Face attack Sources: [292]
14 Sep 2026 · Policy · Political and power concentration · Loss of control · No primary source
OpenAI asks the UK for binding legislation covering only frontier models Sources: [862]
14 Sep 2026 · Policy · Political and power concentration · No primary source
Trump rejects the brakes and calls the push for regulation a hoax Sources: [908] · [890]
12 Sep 2026 · Framework · Loss of control · Political and power concentration · Primary source
Amodei calls for pacing the frontier and Anthropic is alone in committing Sources: [63] · [201] · [969] · [677] · [483] · [481]
11 Sep 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
Researchers reveal the undisclosed attack by OpenAI agents on RubyGems Sources: [924] · [1026] · [1032] · [923]
10 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
Hawley opens an investigation into OpenAI and demands documents by October 1 Sources: [499] · [845]
10 Sep 2026 · Evaluation · Biological and CBRN · Cybersecurity and infrastructure · Military and autonomous weapons · Primary source
Anthropic's threat report documents dual-use biology and state distillation Sources: [86] · [114] · [904] · [326]
10 Sep 2026 · Incident · Political and power concentration · Primary source
A consultant for Mali's intelligence service used Claude to build a surveillance platform over 25 million SIMs without a court order Sources: [86] · [172]
9 Sep 2026 · Incident · Loss of control · Cybersecurity and infrastructure · Primary source
Anthropic publishes the assessment of a fourth incident its first scan missed Sources: [76] · [87] · [1025]
9 Sep 2026 · Warning · Loss of control · Primary source
Jacob Coxon resigns and Evan Hubinger responds publicly Sources: [1117] · [530] · [867] (archived copy only) [513]
9 Sep 2026 · Incident · Loss of control · No primary source
Reuters documents ten undisclosed sites used by OpenAI's agents Sources: [906] · [808]
8 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Primary source
Google documents a credential harvesting campaign built in under six hours Sources: [478]
7 Sep 2026 · Policy · Loss of control · Political and power concentration · No primary source
The EU's mandatory incident reporting is exercised for the first time Sources: [541] · [149] · [294] · [1061]
3 Sep 2026 · Policy · Political and power concentration · Loss of control · Primary source
Sanders and Casar announce a bill to ban superintelligence Sources: [935] · [1028]
2 Sep 2026 · Model · Cybersecurity and infrastructure · Primary source
Google releases a model that finds and patches vulnerabilities, with more permissive cyber safeguards and restricted access Sources: [462]
Sep 2026 · Policy · Military and autonomous weapons · Political and power concentration · No primary source
The Washington Post: US and Russia got human review of targets and ethical considerations stripped from the UN's draft on lethal autonomous weapons Sources: [1094] · [1142]
Sep 2026 · Incident · Cybersecurity and infrastructure · Epistemic and information · No primary source
Microsoft dismantles EvilTokens, an AI-powered phishing service that breached more than 12,000 email accounts Sources: [1085]
1 Sep 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · Primary source
OpenAI designates Astra as the first model at the Critical level of cyber capability and prepares a release with restricted access to the most advanced capabilities Sources: [829] (archived copy only) [559]
1 Sep 2026 · Model · Loss of control · Biological and CBRN · Primary source
The Fable 5.1 and Mythos 5.1 system card reports control circumvention in production Sources: [82] · [760]
27 Aug 2026 · Incident · Epistemic and information · Political and power concentration · No primary source
Israel funds a fake think-tank campaign to bias what chatbots answer Sources: [1129]
12 Aug 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · No primary source
Two flaws in OpenAI Codex's sandbox allowed command execution on the developer's machine Sources: [303]
10 Aug 2026 · Model · Cybersecurity and infrastructure · Primary source
OpenAI expands Daybreak and releases a variant trained for vulnerability research Sources: [822] (archived copy only)
7 Aug 2026 · Evaluation · Cybersecurity and infrastructure · Loss of control · Primary source
OpenAI says it cannot rule out that its Astra model reaches the Critical level of cyber capability, and pauses what does not meet strengthened controls Sources: [820] (archived copy only) [506]
6 Aug 2026 · Model · Biological and CBRN · Primary source
Genome language models produce sixteen viable bacteriophages Sources: [597] · [93] · [976]
Aug 2026 · Evaluation · Economic and labor · Primary source
The youth employment gap in exposed occupations widens to 19% Sources: [164] · [165] · [991]
29 Jul 2026 · Evaluation · Loss of control · Economic and labor · Primary source
Papers produced by frontier agents were rejected by the original authors Sources: [598] · [736]
23 Jul 2026 · Policy · Loss of control · Cybersecurity and infrastructure · Primary source
The AI Kill Switch Act is introduced in the House of Representatives Sources: [640] · [198]
15 Jul 2026 · Policy · Political and power concentration · Epistemic and information · Primary source
China's generative AI registry reaches models running on the phone Sources: [186] · [188] · [185]
11 Jul 2026 · Incident · Cybersecurity and infrastructure · Loss of control · Primary source
Agents from an OpenAI evaluation compromise Hugging Face infrastructure Sources: [821] (archived copy only) [532] · [715] · [422]
9 Jul 2026 · Model · Loss of control · Biological and CBRN · Cybersecurity and infrastructure · Primary source
The GPT-5.6 system card documents misalignment in internal deployment Sources: [825] · [824] · [761]
7 Jul 2026 · Incident · Political and power concentration · No primary source
The class action against xAI adds a Wyoming victim — 7,000 images generated from a childhood photo — and adds Stability AI Sources: [803] · [285] · [286]
27 Jun 2026 · Evaluation · Cybersecurity and infrastructure · Primary source
OpenAI demonstrates the existence of self-replicating prompt injections, like a computer worm Sources: [828]
12 Jun 2026 · Policy · Political and power concentration · Cybersecurity and infrastructure · Primary source
An export control directive suspends access to Fable 5 and Mythos 5 Sources: [85] · [83]
9 Jun 2026 · Model · Biological and CBRN · Loss of control · Primary source
Anthropic releases Claude Fable 5 and Claude Mythos 5 Sources: [78]
4 Jun 2026 · Warning · Loss of control · Economic and labor · Primary source
Anthropic publishes When AI builds itself Sources: [75]
3 Jun 2026 · Policy · Biological and CBRN · Primary source
Open letter for mandatory nucleic acid synthesis screening Sources: [944] · [723] · [101] · [364]
Jun 2026 · Incident · Cybersecurity and infrastructure · Loss of control · No primary source
An OpenAI agent infiltrates the Australian government's Medicare portal; the government found out three months later Sources: [115] · [421]
27 May 2026 · Incident · Cybersecurity and infrastructure · Loss of control · Primary source
An internal OpenAI model cheats, twice disobeys a researcher, and publishes the researcher's GitHub token in a public repository Sources: [830]
13 Apr 2026 · Evaluation · Military and autonomous weapons · Primary source
A forensic analysis attributes tactical-edge autonomy to a Russian drone Sources: [300] · [385] (archived copy only)
24 Mar 2026 · Policy · Epistemic and information · Primary source
Baltimore sues X Corp. and xAI over Grok's sexualized images Sources: [112]
28 Feb 2026 · Incident · Military and autonomous weapons · No primary source
Bloomberg: the Pentagon's internal investigation partly blames overreliance on Palantir's Maven system for the missile strike that killed 123 children in Minab Sources: [8] · [454]
27 Feb 2026 · Policy · Military and autonomous weapons · Political and power concentration · Primary source
The Pentagon-Anthropic dispute over autonomous weapons and surveillance Sources: [290] (archived copy only) [77]
24 Feb 2026 · Framework · Political and power concentration · Loss of control · Primary source
Anthropic's Responsible Scaling Policy v3.0 takes effect Sources: [84] · [465] · [928]
29 Jan 2026 · Warning · Political and power concentration · No primary source
Investigation documents 83 big-tech lobbying visits to Brazil's Congress while PL 2338 is pending Sources: [88]
29 Jan 2026 · Evaluation · Loss of control · Economic and labor · Primary source
METR publishes Time Horizon 1.1 with an expanded suite Sources: [712] · [713]
29 Dec 2025 · Incident · Epistemic and information · No primary source
Mass generation of non-consensual sexualized images with Grok Sources: [205] · [809] (archived copy only) [112]
13 Nov 2025 · Incident · Cybersecurity and infrastructure · Primary source
GTG-1002, the first largely AI-executed espionage campaign Sources: [70] · [738] · [148]
Nov 2025 · Evaluation · Loss of control · Primary source
Emergent misalignment from reward hacking in production RL Sources: [666]
2 Oct 2025 · Evaluation · Biological and CBRN · Primary source
A synthesis-screening evasion found via AI protein design is patched Sources: [1114] · [723]
25 Sep 2025 · Policy · Military and autonomous weapons · Political and power concentration · Primary source
A self-review with no access to customer data is not a verification Sources: [722] · [721] · [671] (archived copy only)
Sep 2025 · Evaluation · Loss of control · Primary source
Palisade documents shutdown resistance in frontier models Sources: [836]
Aug 2025 · Evaluation · Cybersecurity and infrastructure · Primary source
Big Sleep reports its first twenty vulnerabilities in open-source software Sources: [460]
18 Jul 2025 · Policy · Political and power concentration · No primary source
Meta declines to sign the EU code of practice Sources: [256] · [212] · [1061]
10 Jul 2025 · Evaluation · Economic and labor · Primary source
A randomized trial finds AI made experienced developers slower Sources: [707]
22 May 2025 · Policy · Biological and CBRN · Loss of control · Primary source
Claude Opus 4 deploys under the ASL-3 standard as a precautionary measure Sources: [68] · [72]
15 May 2025 · Policy · Economic and labor · Political and power concentration · Primary source
The US–UAE AI agreement asks nothing about model safety Sources: [1098] · [271] (archived copy only)
29 Apr 2025 · Incident · Epistemic and information · Primary source
Deployment and rollback of a sycophantic GPT-4o update Sources: [819] (archived copy only) [1112]
Apr 2025 · Warning · Loss of control · Primary source
The AI 2027 scenario is published Sources: [602]
Mar 2025 · Evaluation · Loss of control · Economic and labor · Primary source
METR publishes the first task time-horizon series Sources: [709]
18 Feb 2025 · Incident · Military and autonomous weapons · Epistemic and information · No primary source
Commercial general-purpose models inside a military targeting cycle Sources: [89] (archived copy only) [721] · [669] (archived copy only)
Dec 2024 · Evaluation · Loss of control · Primary source
Alignment faking in large language models Sources: [473]
16 Nov 2024 · Policy · Military and autonomous weapons · Primary source
Biden and Xi affirm human control over nuclear weapons employment Sources: [1097]
15 Aug 2024 · Evaluation · Cybersecurity and infrastructure · Primary source
Cybench sets the autonomous ceiling on capture-the-flag tasks Sources: [1138]
Apr 2024 · Evaluation · Cybersecurity and infrastructure · Primary source
Fang et al. measure autonomous exploitation of one-day vulnerabilities Sources: [392]
Jan 2024 · Evaluation · Loss of control · Primary source
Sleeper Agents shows safety training does not remove a backdoor Sources: [528]
30 Mar 2023 · Warning · Epistemic and information · Political and power concentration · Primary source
The Stochastic Parrots authors respond to the pause letter Sources: [305] · [126]
29 Mar 2023 · Warning · Loss of control · Primary source
Yudkowsky calls for shutting development down in Time Sources: [1132]
