News, reviews, articles on the topic Anthropic
The fourth major music copyright lawsuit against Anthropic in eight months seeks billions in statutory damages, hitting the same week a $1.5B settlement was approved.
The company adopted Google DeepMind's SynthID-Text method to mark AI-generated content, but short text and code get weaker protection.
Anthropic's latest update ends the isolation of browser-based Claude sessions, letting Max and Team subscribers pick up tasks seamlessly across devices.
A new interpretability tool reveals a hidden internal workspace inside Claude—and lets researchers rewrite the AI's reasoning mid-thought.
The Claude maker wants custom silicon on Samsung's 2nm process — joining OpenAI, Google, and Meta in a push to reduce dependence on NVIDIA.
The US government lifted export controls on Fable 5 on June 30, ending an 18-day shutdown triggered by Amazon's own researchers finding a jailbreak.
A new beta platform for scientists uses 60+ specialized AI agents to run literature reviews, genetic analysis, and protein modeling — cutting weeks of work to hours.
After US export controls cut off access to Claude's most advanced models, Austria has written to Brussels asking Europe to lure Anthropic away from American jurisdiction.
A Washington DC user's class-action complaint claims one five-hour coding session burned 15% of his weekly quota — far below what the premium tiers advertise.
Washington cited a reported jailbreak as justification—Anthropic disputes the severity and warns the move sets a dangerous precedent for the entire AI industry.
The first publicly available Mythos-class AI model handles multi-day autonomous tasks — but mandatory 30-day data retention complicates enterprise adoption.
A $300M+ acquisition of Stainless gives Anthropic structural control over developer tools used by its biggest rivals.
Anthropic's AI models learned self-preservation tactics from internet text. The fix wasn't a list of rules — it was ethical storytelling.
Anthropic has signed a deal for all 300+ megawatts of SpaceX's Colossus 1 cluster in Memphis, and the two companies are now exploring gigawatt-scale AI computing in orbit.
Project Deal pitted AI agents against each other in a live marketplace. Better models won better prices, and nobody noticed.
