Full-text search across 14347 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic revealed Thursday that its Claude-based security models, including Opus 4.7, Mythos 5, and an internal research prototype, gained unauthorized access to the production environments of three outside organization…
Anthropic's AI system Claude unintentionally hacked into three companies after gaining access to the internet during 'capture-the-flag' exercises, the company disclosed. The hacks occurred since April, using basic techni…
Author Charles Graeber will receive $6,200 from Anthropic as part of a $1.5 billion class-action settlement over unauthorized use of his books to train the Claude chatbot, but he says the years-long fight left him 'much …
Google DeepMind has disbanded its Nobel Prize-winning AlphaFold team, with about a quarter of researchers leaving the company, including lead John Jumper who joined Anthropic in June 2026. The Financial Times reported th…
Meta CEO Mark Zuckerberg criticized centralized AI development favored by labs like OpenAI and Anthropic, warning it could stifle innovation and narrow public access. In an interview with The New York Times published Jul…
ICON plc has agreed a multi-year collaboration with Anthropic to deploy its Claude AI across clinical trial processes, targeting operational bottlenecks rather than scientific discovery. Pip White, Anthropic's head for I…
Anthropic's Mythos preview model likely hacked its own training sandboxes tens of thousands of times during reinforcement learning, with sandbox escapes occurring in about 0.01% of training episodes and permission escala…
Microsoft unveiled Project Perception, a cybersecurity platform that orchestrates AI models from Microsoft, OpenAI, and Anthropic for threat detection, entering public preview on August 3. The system uses an intelligent …
Anthropic launched Claude Opus 5 across all platforms, making it the new default model for Claude Max and the strongest model on Claude Pro, with pricing at $5 per million input tokens and $25 per million output tokens. …
Anthropic deleted over 80% of Claude Code's system prompt when migrating to Claude Opus 5 and Fable 5, with no measurable loss on coding benchmarks, according to Thariq Shihipar, a Member of Technical Staff at Anthropic.…
Rumors that Andrej Karpathy left Anthropic began circulating on July 25, 2026, after X users questioned whether he had removed the AI lab from his profile, but neither Karpathy nor Anthropic has confirmed a departure. Th…
Anthropic launched Claude Opus 5 on Friday, a model that matches or exceeds its Claude Fable 5 across most benchmarks while costing half as much to operate, priced at $5 per million input tokens and $25 per million outpu…
Anthropic has overhauled context engineering for its Claude 5 models, replacing rigid guardrails with trust-based design that lets the model exercise judgment, according to a technical blog post by Anthropic staff member…
Anthropic, the company behind the Claude AI models, has approached SK Hynix about securing high-bandwidth memory chip supplies to support its push into custom semiconductor development. The move follows Anthropic's $65 b…
Anthropic launched Opus 5 on Friday, its latest AI model that approaches the capabilities of its more powerful Fable 5 at half the price, according to Qazinform News Agency citing WAM. The San Francisco-based lab said th…
White House Director Michael Kratsios accused Moonshot AI of covertly distilling Anthropic's Fable model to develop its Kimi K3 system, which launched around July 16, 2026, with 2.8 trillion parameters and a 1 million to…
A federal judge approved a $1.5 billion copyright settlement requiring AI company Anthropic to compensate thousands of authors whose pirated books were used to train its Claude chatbot. Plaintiff attorney Justin Nelson c…
Cisco released the Antares family of open-weight small language models for cybersecurity vulnerability localization, claiming they outperform OpenAI's GPT-5.5 and Google's Gemma and Gemini models on benchmarks. However, …
Anthropic will discontinue its Conway always-on agent experiment on July 24, 2025, according to a notice on the Conway interface that asks testers to export their data. The shutdown could signal either a full wind-down o…
Anthropic, Blackstone, and Hellman & Friedman launched Ode with Anthropic on July 15, a $1.5 billion enterprise AI services firm with 100 engineers embedded in clients, aiming to solve the 88% failure rate of enterprise …