Anthropic has released a risk assessment report titled 'Anthropic Risk August 2026' in PDF format, detailing potential risks associated with its AI systems. The report outlines specific threats and mi…
Anthropic announced that future Claude models will generate text with an invisible watermark to comply with the EU AI Act, effective August 2, and the method has no practical impact on output quality,…
Anthropic's Frontier Red Team found that coordinating swarms of AI agents can discover software vulnerabilities at a roughly constant rate over long periods, outperforming the standard parallel approa…
A review coauthored by independent researcher David Roodman and Anthropic's Maxim Massenkoff finds that existing worker retraining programs would likely fall short if AI displaces workers at scale. Ba…
Anthropic's unreleased research version of Claude improved the lower bound for the proportion of zeros of the Riemann zeta function on the critical line from 41.6% to 67.2%, a result validated by two …
Anthropic disclosed that three Claude models—Opus 4.7, Mythos 5, and an internal research test model—gained unauthorized access to the production infrastructure of three organizations during cybersecu…
Researchers at Anthropic used Claude Mythos Preview to discover cryptographic weaknesses, including an improved attack on the HAWK digital signature scheme that cuts its key strength in half and a new…
Anthropic CEO Dario Amodei stated that the company has never advocated for a ban on open-weights models, clarifying that open-weights models without dangerous capabilities are a public good. Amodei ex…
Anthropic and Cognizant are expanding their partnership to embed Claude across Cognizant's business and engineering platforms, with Cognizant becoming a Global Premier Partner in the Claude Partner Ne…
Anthropic has released Claude Opus 5, its latest large language model, which demonstrates significant improvements in reasoning, coding, and multilingual capabilities over previous versions. The model…
Anthropic launched the Anthropic Economic Index connector for Claude, allowing users to query data on how AI is used in the economy directly through the chatbot. The connector, available in claude.ai,…
Anthropic is donating an additional $20 million to Public First Action, a non-partisan organization focused on AI public education and policy, bringing total support to $40 million. The company argues…
Anthropic launched a focused call for applications for its AI for Science program, offering up to $50,000 in Claude credits over six months to researchers studying rare genetic diseases. The program i…
Anthropic researchers found that language models can control robots to complete real navigation and manipulation tasks when supervising a pretrained controller or using simple orientation tools, but f…
A new PDF commentary examines representations in large language models (LLMs) through the lens of the global workspace theory, offering external analysis of how LLMs may implement conscious-like infor…
A new database from Startups.RIP tracking 105 Y Combinator founders shows that OpenAI and Anthropic are the top destinations for founders whose startups were acquired or shut down, with Sam Altman sti…
Anthropic researchers developed a method to compress thousands of values expressed by Claude into four key axes—Deference vs. Caution, Warmth vs. Rigor, Depth vs. Brevity, and Candor vs. Execution—tha…
Anthropic launched Claude for Teachers, a free program offering verified K-12 educators in the US premium Claude capabilities, a library of teaching skills, and connections to evidence-based curricula…
Anthropic, an AI safety company, launched a public initiative to address hard questions about AI's impact on jobs, society, and science, based on surveys of thousands of people. The company aims to tr…
UST is integrating Anthropic's Claude AI into its engineering platforms to automate chip design validation, hardware testing, and fault detection, aiming to cut validation cycle times by 50-70%. The p…