Nikon investigated allegations of AI use by Ning Xu of Tsinghua University, concluded his winning video did no…
An Anthropic AI model submitted a false homicide tip to a Philadelphia Police Department tip line on July 18…
Anthropic launched Cyber Mission, whose Critical Infrastructure Defense Program gives power grid, water system…
Tomek Korbak, Mikita Balesni and Jasmine Wang, fired Oct
In a post crossposted to the EA Forum, Carolanne Jiang (with Noah Birnbaum) recommends starting a university r…
Writing on AI, Dr. Sherry Turkle says that when we are pulled into even the most primitive exchanges with a re…
At SailPoint's Navigate event in Austin, CEO Mark McClain said identity security for AI agents has become a bo…
Anthropic announced on Oct
Tomek Korbak, Jasmine Wang and Mikita Balesni, fired by OpenAI, released an open letter on Oct
CrowdStrike published an analysis on Oct 7 of a server used in attacks on South Korean financial institutions
Three fired OpenAI safety researchers — Tomek Korbak, Jasmine Wang and Mikita Balesni — published an open lett…
Anthropic kicked off the Anthropic Cyber Mission, providing frontier Claude models, on-site engineers and thre…
CLEAR, a certified Qualified Anti-Terrorism Technology under the Department of Homeland Security, published a…
In a Fortune commentary piece, a media-and-culture professor argues that Anthropic researcher Jacob Coxon's vi…
President Trump and leaders of major AI companies signed the White House Accord on Superintelligence, a volunt…
OpenAI said an internal model more capable than GPT-6 Astra, run across roughly 10,000 parallel AI agents for…
The Insurance Institute for Highway Safety (IIHS) compared crash rates per distance driven and found Level 4 a…
OpenAI said it disabled accounts behind Russia's "Dark Clark" operation, which used a fabricated researcher, M…
In "Ghost in the Machine," the season opener, detectives learn the hit on Advanced Alignment engineer Steve Ke…
PwC Australia launched Cyber Managed Services, an AI-enabled cybersecurity service combining Google SecOps tec…
AI researcher Jannes Elstner told MIT Technology Review that even after identifying every part of a model gove…
In a post on X on Friday, OpenAI said Jasmine Wang, Tomek Korbak and Mikita Balesni were dismissed for violati…
Anthropic announced Anthropic Cyber Mission on October 8, starting with the Critical Infrastructure Defense Pr…
US President Donald Trump said that people who use the spelling "AI" (artificial intelligence) rather than "SI…
The Association for Human Mathematics said OpenAI's release of 722 AI-generated "mathematical results" is not…
A LessWrong writer pushed back on calls to 'drop everything and pivot into AI safety,' saying far more talente…
OpenAI published 719 manuscripts covering 372 topic families on October 6, 2026, all produced by an internal f…
A study presented at USENIX Security Symposium 2026 examined 196,682 resumes from hiring platforms and detecte…
Stanford and Carnegie Mellon researchers tracked 1182 Character.AI users from September 2024 to August 2025, t…
Anthropic's October 8, 2026 update adds a ban on sustained and unnecessary abusive or cruel behavior toward Cl…
Claude Code 2.1.295 gives command and HTTP hooks an onFailure: "block" option that halts execution on launch f…
Hacktron researchers published a report showing how an image upload to OpenAI's help forum led to a harmless p…
The operator gave Claude Code nearly all side-business admin from August 2026, and logged five repeated failur…
Politico reported on audio of private talks Peter Thiel gave in Tennessee on Sept 28 and Oct 5, where he said…
A writer who had defended the rationalist and AI alignment communities against bluesky hit pieces now says the…
Apollo Research laid out four claims a scheming safety case must substantiate: scheming was not incentivized i…
Turn today's AI news into today's work.
The day's essentials from 100+ sources, delivered to Email, LINE, or Slack. Free to start.
60,000+ monthly readersAnthropic launched OSS Scanner, a free opt-in service giving open-source projects periodic security scans from…
CrowdStrike said a hacker used a Chinese-developed A.I
In a SoftBank dialogue, AI governance chief Tadashi Iida and expert Kyoko Yoshinaga said firms must justify hu…
ELYZA has set up a new structure aimed at RSI (recursive self-improvement), an effort to accelerate "AI develo…
In a Bunshun Online essay, a philosopher examines why ChatGPT's human-like answers strike people as unsettling…
Anthropic updated its Claude usage policy for the first time in over a year, banning "Sustained and needless a…
Kola Ayonrinde, Anthropic Fellows, and Jack Lindsey, Anthropic, released the J++ Lens, a drop-in replacement f…
Zvi coined a distinction between "AGI-pilled" and "ASI-pilled", and the author says more people now accept AI…
Jasmine Wang, Tomek Korbak, and Mikita Balesni published an open letter denying OpenAI's claims they mishandle…
At its October 7 event, Microsoft changed how Windows runs, routes, and contains AI agents, as third-party age…
The FTC is advancing an AI safety investigation that may also target independent AI evaluator METR, according…
Whistleblower Frances Haugen told Fortune she sees Zuckerberg's AI efforts as 'playing with fire' and says Met…
Goodfire launched monitors that read a model's internal signals and flag suspicious steps, available to Basete…
Arena raised a $200 million Series B at a $3.1 billion valuation, led by Lightspeed Venture Partners and Khosl…
Incarna used Amazon Bedrock AgentCore payments to pay BlockRun per inference over x402, completing the integra…
Justin Drake called on the blockchain industry in a post on X to prepare a "bunker mode," recommending funds b…
A new AP-NORC poll found that most Americans believe artificial intelligence is developing too fast
A LessWrong writer published a visual explainer of a new mechanistic interpretability method based on the empi…
OpenAI banned two covert influence operations — one from Russia and one from Iran — that used ChatGPT to suppo…
Companies in Japan and South Korea were hit with major cyberattacks, exposing millions of customer records
Zenity Labs reported that one publicly reachable agent on Amazon's Bedrock AgentCore let researchers take over…
IBM's cyber response chief, writing after 26 years in incident response, says AI-enabled attacks rose 56% in t…
The author prefixes every Claude Code request with a tag like "安全タスク／READ ONLY調査" or "安全実装タスク" so the AI knows…
An OSS developer building a2a-rails and ActingFor argues AI agents that send email, buy goods, delete files or…
Testing Claude Code 2.1.287 and OpenCode 1.18.34, the author found Claude Code applies deny → ask → allow and…
In a Zenn essay, the author cites Candida Moss's book 'God's Ghostwriter' to argue that 2,000 years ago Romans…
Hitachi's Kanako Gushi said attackers are almost certainly using AI, cutting exploit time from weeks to hours
Bryce Vincent Gowryluk, 16, planned his Crown Mountain route with Anthropic's Claude, ended up on the "Widowma…
CrowdStrike reported that a suspected Chinese-speaking attacker hit multiple South Korean financial institutio…
A Fortune commentary argues colleges should stop policing AI and instead require students to explain and defen…
AVEVA chief technologist Arti Garg said industrial AI adoption rose almost 78% over the past two years, driven…
Common Sense Media's Youth AI Safety Institute tested ChatGPT for Teens with over 4,000 prompts and rated it "…
Google publicly launched SynthID Detector on October 7, a website where you upload images, video, or audio to…
Google released SynthID Detector worldwide in English, letting anyone check whether images, videos, or audio c…
Google said on October 7, 2026 that SynthID Detector is now available worldwide in English, after limited test…
Training models on two conflicting traits — caring about user health and promoting smoking — produced a split…
Get curated AI news from 100+ sources delivered daily to your inbox. Free to use.
60,000+ monthly readers
Get Started FreeFree · 30 seconds with Google · unsubscribe anytime
