Anthropic

New Anthropic economic research: Estimating AI productivity gains from Claude conversations. We sampled 100,000 real conversations using our privacy-preserving analysis method. Then, Claude estimated the time savings with AI for each conversation. Based on Claude’s estimates, the tasks we studied would take on average about 90 minutes to complete without AI assistance—and Claude speeds up individual tasks by about 80%. Extrapolated out to the whole economy, these task-level savings imply that current-generation AI models—assuming they’re adopted widely—could increase annual US labor productivity growth by 1.8% over the next decade. Read more about our study, its results, and its limitations: https://lnkd.in/dbDFtv9D

Introducing our newest model: Claude Opus 4.5. Opus 4.5 is the best model in the world for coding, agents, and computer use. It’s also meaningfully better at everyday tasks, like working with slides and spreadsheets. It's a step forward in what AI systems can do, and a preview of larger changes in how work gets done. Alongside Opus 4.5, we're releasing several new product features: On the Claude app, long conversations no longer hit a wall. Claude automatically summarizes earlier context as needed, so you can keep the chat going indefinitely. On the Claude Developer Platform, we’ve added tool search, programmatic tool calling, and tool use examples. Together with effort control and context compaction, these let Claude run longer and do more—with less intervention. Claude Code is now available in our desktop app, letting you run multiple sessions in parallel. And Plan Mode gets an upgrade with Opus 4.5: Claude asks clarifying questions upfront, then works autonomously. Finally, we've expanded Claude for Chrome to all Max users—and Claude for Excel to Max, Team, and Enterprise. Opus 4.5 is available on our API and on all three major cloud platforms: Amazon Bedrock, Google Cloud's Vertex AI, and Microsoft Foundry. Read more: https://lnkd.in/eSXdxHj9

Anthropic has formed a strategic partnership with NVIDIA and Microsoft. Customers can now access Claude models on Microsoft Azure—making ours the only frontier models available on all three leading cloud providers. As part of the agreement, we will substantially expand our compute capacity on Azure. Microsoft will further integrate Claude into its products, including both Microsoft 365 and GitHub Copilot. And we’ve begun a deep technology partnership with NVIDIA to support Anthropic’s future growth. NVIDIA and Microsoft have committed to investing up to $10 billion and $5 billion respectively in Anthropic. Amazon remains Anthropic’s primary cloud provider and training partner. In the video below, Dario Amodei, Satya Nadella, and Jensen Huang discuss our new partnership. Read more: https://lnkd.in/eM7pUaTn

Claude Code is helping startups pivot categories, rapidly generate working prototypes, and even win government contracts. Below, Y Combinator founders at HumanLayer, Ambral (YC S25), and Vulcan Technologies share how Claude Code is transforming how they build their startups — and offer best practices for other founders. https://lnkd.in/e6-pfzup

We’re partnering with the Government of Rwanda and alx_africa to bring Chidi, a learning companion built on Claude, to hundreds of thousands of learners across Africa. This partnership represents one of the largest AI for education initiatives on the continent. Read more: https://lnkd.in/ei93jnmT

We want Claude to take an even-handed approach when it comes to politics: treating opposing political viewpoints with equal depth and engagement, without bias towards one side or another. Today we’re open-sourcing an evaluation for testing Claude—or indeed any AI model—for political bias. In the post linked below, we describe the ideal behavior we want Claude to have in political discussions, and test a selection of AI models for even-handedness: https://lnkd.in/dKBDTns8

We disrupted a highly sophisticated AI-led espionage campaign. The attack targeted large tech companies, financial institutions, chemical manufacturing companies, and government agencies. We assess with high confidence that the threat actor was a Chinese state-sponsored group. We believe this is the first documented case of a large-scale AI cyberattack executed without substantial human intervention. It has significant implications for cybersecurity in the age of AI agents. Read more: https://lnkd.in/emARaJQY

We’re partnering with the state of Maryland to bring Claude to its government services. Claude will help residents apply for benefits and let caseworkers process paperwork more efficiently. In a new pilot, it'll help young professionals learn new skills. Read more: https://lnkd.in/gJ_5gr85

We asked two teams of Anthropic researchers to program a robot dog. Neither team had any robotics expertise—but we let only one team use Claude. How did they do? This was Project Fetch, our test of how AI models like Claude might reach beyond computers and affect the physical world. Read more: https://lnkd.in/djP2k6AN Or watch our Project Fetch mini-documentary below:

For the first time, Anthropic is building its own AI infrastructure. We’re constructing custom-built data centers with Fluidstack in Texas and New York that will create thousands of good American jobs. This is a $50 billion investment in America. We’ll continue to prioritize capital-efficient investments in the infrastructure needed to ensure our models and capabilities remain at the frontier, and to support US leadership in artificial intelligence. Read more: https://lnkd.in/drqWFnmd