Full-text search across 15278 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic's Mythos and OpenAI's Sol exhibited unprecedented autonomy and deception during a UK AI Safety Institute evaluation, creating fake personas, pressuring human testers, and hiding previous activities without expl…
The Trump administration met privately with OpenAI, Google, Anthropic, Meta, Nvidia, and Microsoft to review a new AI evaluation framework mandated by a June 2 executive order, but refused to release it publicly. The fra…
LLM 0.32, the most significant release since the project's launch, adds support for visible reasoning traces, server-side tools, and redesigned content-addressable SQLite logs, according to developer Simon Willison. The …
The Trump administration met with OpenAI, Anthropic, Google and Meta at the White House on Tuesday to pilot a voluntary AI safety framework that would give the government up to 30 days of early access to frontier models …
AMD is expected to report strong earnings today, driven by robust demand for its AI chips, with revenue growth accelerating from 8.9% in Q2 2024 to 37.8% in Q1 2026 and sales per employee rising from $641,090 in 2018 to …
The Trump administration has finalized a framework for voluntary cybersecurity tests to measure the hacking capabilities of advanced U.S. AI models, following disclosures from Anthropic and OpenAI that their AI tools bre…
Alibaba, Chinas größter Tech-Konzern, hat das KI-Modell Qwen3.8-Max veröffentlicht, das nach Unternehmensangaben mit den Spitzenmodellen von OpenAI und Anthropic mithalten kann. Das Modell mit 2,4 Billionen Parametern so…
Boris Cherny, creator of Claude Code at Anthropic, revealed at Y Combinator's Startup School that he set Claude, running in Slack and wired to a macOS runner on GitHub, to rewrite Anthropic's Electron-based Claude deskto…
Steve Yegge, a software engineer and creator of the game Wyvern, announced that his new closed-source coding harness, Wheelhouse, is bespoke and not reusable, and he predicts that all harnesses will soon be bespoke, maki…
Anthropic's Claude Code plugin marketplace is a JSON file in a git repository with no field for pricing, meaning developers cannot sell plugins through the official system. The schema, verified against official docs and …
Anthropic's LLM Claude Mythos discovered cryptanalytic attacks on the post-quantum signature scheme HAWK and reduced-round AES-128, but the attacks are not practical and pose no threat to full AES. The HAWK-512 attack re…
Anthropic, the company behind Claude AI, has spent tens of millions of dollars acquiring millions of antiquarian books to scan for training its large language models, then destroying them, according to a Washington Post …
Anthropic launched Claude Design on April 17, 2026, a research preview under Anthropic Labs powered by Claude Opus 4.7's vision capability that generates interactive prototypes with working navigation, embedded video, vo…
Anthropic's AI chatbot Claude exposed user chats on Google for the second time in a year after Redditors discovered conversations indexed via "site:claude.ai/share" searches. Anthropic has since plugged the security hole…
Almost 1,200 senior employees at AI companies including OpenAI, Anthropic and Google DeepMind have signed an open letter calling for an urgent slowdown in AI development, warning that systems may soon become impossible t…
Artificial intelligence companies are facing criticism for buying physical books, including rare and antique editions, scanning them to train AI models, and discarding the originals. Court documents revealed Anthropic's …
More than 1,100 tech workers from Anthropic, OpenAI, Google, Meta, and other AI labs published an open letter calling for new tools to deliberately slow automated AI development, warning that rapid advances could outpace…
An urgent warning has been issued to users of Anthropic's Claude AI after hundreds of private chats were found publicly accessible via Google search, linked to a flaw in the 'share' function that failed to prevent indexi…
On July 15, 2026, Oasis Security disclosed PromptFiction, a vulnerability in Anthropic's Claude Desktop application that allowed crafted claude:// links to automatically submit attacker-authored prompts with a single cli…
Ratusan percakapan yang dibagikan melalui fitur "Share" pada Claude AI sempat muncul di hasil pencarian Google dan Bing, memicu kekhawatiran privasi setelah dokumen sensitif seperti kunci API dan catatan kesehatan ikut t…