Anthropic — Latest News and Analysis

Alphabet’s data-center agreements could require it to cover a tenant’s payments after a default, with rights to use or sublet the facility.

AI labs DeepSeek and Moonshot AI reportedly trained on outputs from OpenAI's Astra model for their latest releases.

OpenAI is rolling out phone messaging support for Dot, its always on task agent, so ChatGPT Pro users can send requests and get progress updates in their phone’s messaging app.

The new Artificial Analysis’ Intelligence Index v5 bases 60% of its score on private tests and adds new coding and science tasks. No single AI model leads them all.

Generate Biomedicines used Anthropic's Claude AI to help redesign receptors on immune cells which kept killing cancer cells longer than usual in lab tests.

Agent Skills lets teams reuse instructions across compatible AI agents. The format has joined the Agentic AI Foundation, with shared governance planned.

Anthropic’s pending IPO will give founders 50.1% of general shareholder votes, while a separate trust would choose most directors.

Monhot AI's Kimi K3.1 Pro model scores higher than its predecessor and than GPT-6 Astra in AI benchmarks, while extra reasoning and tool calls can offset its lower token rates.

OpenAI examined the phones of their three recently fired safety researchers. Two say talks with an outside researcher were part of their jobs; OpenAI alleges broader misconduct.

DeepSeek's V4.1 Flash 1017 update improves on coding scores and lowers task costs, while factual reliability still trails Claude Opus 5.5.

Harm warnings did not reliably stop Mythos 5 AI’s simulated attacks. Newer Anthropic models showed less harmful behavior, but the cause remains disputed.

Password manager1Password is rolling out user-approved website logins for more AI agents. It can apply previously saved user passwords without giving them to agents directly.

OpenAI is reportedly seeking about $30 billion in investment commitments before rival Anthropic’s expected November stock-market debut, as both court some of the same backers.

AI agents working for a hacker reportedly spent hours breaking into more than 100 company websites.

Anthropic and OpenAI are reportedly negotiating what outside AI safety reviewers can publish. The terms could shape what the public learns about AI risks.

Anthropic's new Claude Haiku 5.5 small AI model adds adjustable reasoning for summaries and delegated tasks and beats OpenAI's cheap GPT-6 Luna model in benchmarks.

Word choices will carry hidden OpenAI watermarks in ChatGPT text responses for EU users.

Eight former OpenAI, Anthropic and Google DeepMind employees describe different reasons for leaving, including limits on speaking publicly and choosing their research.

Claude for Government’s desktop AI assistant is generally available to eligible US federal and state agencies, helping staff work with files and develop software.

The Federal Trade Commission is investigating potential consumer risks from OpenAI and Anthropic’s AI products. Formal demands for records and executive testimony are being prepared.

Google's Gemini 4 Argon AI is reaching selected cyber defenders. It can generate longer text in one run; ordinary developers and subscribers have to wait for access.

The lobby groups with close ties to AI labs "Leading the Future" and "Public First" fund candidate ads across party lines to influence who writes AI rules in Congress.

OpenAI has released its GPT-6.1 Sol AI model. In Artificial Analysis's independent ten-test index, it scored 52 versus GPT-6 Sol's 48 with substantially lower measured cost per task.

Anthropic has released its Claude Sonnet 5.5 AI model at Sonnet 5's price. Anthropic says some tasks cost less; Artificial Analysis' max-effort test found higher task cost than Sonnet 5.

At a U.N. Security Council briefing, OpenAI and Anthropic CEOs urged shared AI model tests and serious-incident alerts. The Council adopted no AI rule, and the U.S. rejected a global regulator.

Anthropic released Claude Opus 5.5 as OpenAI introduced GPT-6 Sol and Luna on September 22, bringing lower API prices to three new models.

Anthropic's possible November IPO could fund computing for model training and customer use of its Claude AI assistant, while CEO Dario Amodei argues for slower advances in AI capabilities.

Anthropic proposes ongoing access for outside reviewers to inspect AI development and publish findings, making its promises checkable; slower capability growth across labs requires coordination.

Google Cloud and Accenture have launched a joint group to turn AI pilots into working business systems, backed by a planned 1,000-person engineering workforce.

OpenAI claims its AI agents solved the Navier-Stokes Equations about mathematically expressed momentum balance for Newtonian fluids,, while mathematician Tristan Buckmaster challenges its handling of already available research.