Loading…
Anthropic
AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
- 187 posts summarised
- latest
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
From their blog
Latest articles
Anthropic expands to India with Bengaluru office
Anthropic announces plans to open a Bengaluru office in early 2026, its second Asia-Pacific location after Tokyo, to serve India’s growing AI ecosystem and rising demand for Claude. CEO and co-founder Dario Amodei is visiting India to meet public officials and enterprise partners, while the company says its regional work will target social impact in education, healthcare, and agriculture and support enterprises, nonprofits, startups, and the entrepreneurial ecosystem. India ranks second globally in consumer Claude usage, with substantial use for programming tasks; Anthropic says Claude Code usage grew more than 10x in its first three months after launch. The expansion also includes enhanced Hindi performance and support improvements for nearly a dozen other Indic languages, intended to broaden public-sector adoption and access.
Salesforce and Anthropic expand partnership
Salesforce and Anthropic announced an expanded partnership making Claude a preferred model for Salesforce’s Agentforce platform, with availability aimed at customers in regulated sectors including financial services, healthcare, cybersecurity, and life sciences. Through Amazon Bedrock, Claude traffic will remain within Salesforce’s virtual private cloud and trusted environment, while industry-specific solutions are being developed beginning with financial services. The companies are also deepening Claude’s Slack integration through a Model Context Protocol server that can access channels, messages, and files for summaries, decisions, drafts, searches, and connected Salesforce CRM and Tableau insights. Salesforce is deploying Claude Code across its global engineering organization, and Anthropic uses Claude in Slack for internal sales workflows. Agentforce powered by Anthropic and the Claude-Slack bidirectional integrations are available today, while the regulated industry collaboration and Agentforce 360 integration for Claude remain under development.
Dario Amodei on American AI leadership
Anthropic CEO Dario Amodei says the company’s aim is to advance American leadership in powerful, beneficial AI while maximizing helpful applications and limiting harmful ones. He describes cooperation with the federal government, support for a national AI standard, and support for California’s SB 53, which requires the largest AI developers to publish frontier-model safety protocols and exempts companies below $500 million in annual gross revenue. Amodei also says Anthropic restricts AI-service sales to PRC-controlled companies, works with startups, and has hired advisers from both political parties. Addressing claims of unusual political bias, he cites Manhattan Institute and Stanford findings and Anthropic system cards as evidence that its models are not uniquely biased, while acknowledging that no model is fully balanced in every reply.
Seoul becomes Anthropic's third APAC office
Anthropic plans to open an office in Seoul in early 2026, making it the company’s third Asia-Pacific location after Tokyo and Bengaluru as it expands operations in Korea. Korean users rank among Claude’s top five globally for total and per-capita usage, while Asia-Pacific countries account for more than a quarter of Claude Code users and weekly users in Korea have grown sixfold in four months. Anthropic attributes the expansion to regional momentum, including more than tenfold run-rate revenue growth in the past year and an eightfold increase in large Asia-Pacific business accounts; it also points to Korean deployments such as Law&Company’s legal assistant and SK Telecom’s customer service model. The Seoul team will support Korean enterprises and startups, work with government, research, and industry partners, and align with Korea’s strategy for responsible AI development and deployment.
Expanding our use of Google Cloud TPUs
Anthropic announces plans to expand its use of Google Cloud technologies, including up to one million TPUs, in an expansion worth tens of billions of dollars. The company expects the deployment to bring well over one gigawatt of capacity online in 2026, increasing resources for customer demand, testing, alignment research, and responsible deployment at scale. Anthropic says it serves more than 300,000 business customers and that large accounts, defined as customers representing more than $100,000 in run-rate revenue, have grown nearly sevenfold in the past year. The company describes a diversified compute strategy using Google TPUs, Amazon Trainium, and NVIDIA GPUs, while remaining committed to Amazon as its primary training partner and cloud provider and continuing Project Rainier.
Advancing Claude for Financial Services
Anthropic is expanding Claude for Financial Services with a beta Excel add-in, real-time data connectors, and six pre-built Agent Skills for finance workflows. Claude for Excel operates in a Microsoft Excel sidebar, where it can read, analyze, modify, debug, and create workbooks while tracking changes, explaining actions, and linking explanations to referenced cells. New connectors provide live or governed access to earnings events, portfolio data, internal data rooms, market data, credit research, and financial news through Aiera, Chronograph, Egnyte, LSEG, Moody’s, and MT Newswires. The six skills cover comparable company analysis, discounted cash flow models with WACC, scenarios, and sensitivity tables, due-diligence data packs, earnings analysis, and initiating coverage reports. Sonnet 4.5 scored 55.3% on Vals AI’s Finance Agent benchmark; these previews are available to Max, Enterprise, and Teams users, while Claude for Excel will gather feedback from 1,000 initial users.
Anthropic opens Tokyo office
Anthropic opened its first Asia-Pacific office in Tokyo as part of its international expansion, building a local team to work with industry, government, and cultural partners. During the week, CEO and co-founder Dario Amodei met Japan’s prime minister, the LDP Digitization Headquarters Committee, and customers, while Anthropic signed a Memorandum of Cooperation with the Japan AI Safety Institute on AI evaluation methodologies and emerging trends. Anthropic also joined the Hiroshima AI Process Friends Group and extended its partnership with the Mori Art Museum. The company reports that Japan is in the top 25% globally for AI adoption, with businesses using Claude for research, writing, document analysis, coding, and other productivity tasks. It plans to expand next to Seoul and Bengaluru, while reporting Asia-Pacific run-rate revenue growth of more than 10x in the past year.
Cognizant brings Claude to 350,000 employees
Cognizant announced plans to deploy Anthropic’s Claude to up to 350,000 employees globally, using it with agentic tooling, engineering platforms, and industry blueprints to move enterprise AI efforts from experimentation to production outcomes. It will make Claude available across corporate functions, engineering, and delivery teams, while Claude Code supports coding, testing, documentation, and DevOps workflows. The rollout also aligns Claude models, Claude Code, the Model Context Protocol (MCP), and the Agent SDK with Cognizant’s platforms for integration with existing data and applications, multi-step work, human oversight, and management of performance, risk, and spend. The stated applications include software productivity through Cognizant Flowsource, legacy modernization, multi-agent systems via Cognizant Neuro and Agent SDK, and regulated industry solutions beginning with financial services. Cognizant also describes responsible AI practices for deployment, monitoring, operations, and governance, and plans client workshops, reference patterns, and platform integrations to identify use cases and move them into production.
New offices in Paris and Munich
Anthropic announces plans to open offices in Paris and Munich as it expands operations across Europe. The offices will serve as regional hubs for commercial growth, policy collaboration, and partnership development, with European staff spanning research and engineering through sales and operations. The expansion follows a year in which EMEA employee numbers tripled, run-rate revenue grew more than 9x, and large business accounts grew more than 10x; EMEA is described as Anthropic’s fastest-growing region. New leadership roles will cover startups and digital-native businesses, northern and southern EMEA, and, pending an announcement, DACH and CEE, while partnerships with European educational and cultural organizations will support exhibitions, hackathons, and AI education.
Anthropic invests $50 billion in American AI infrastructure
Anthropic announces a $50 billion investment in American computing infrastructure, with data centers being built by Fluidstack in Texas and New York and additional sites planned. The facilities are custom built for Anthropic, with efficiency for its workloads intended to support continued frontier research and development. Sites are expected to come online throughout 2026, creating approximately 800 permanent jobs and 2,400 construction jobs, while the project is described as supporting goals in the Trump administration’s AI Action Plan and domestic technology infrastructure. Anthropic says the investment is needed to meet demand from hundreds of thousands of businesses using Claude; it reports more than 300,000 business customers and nearly sevenfold growth in large accounts over the past year. Fluidstack was selected for its ability to deliver gigawatts of power rapidly, and Anthropic says it will continue pursuing cost-effective, capital-efficient ways to scale.
Measuring political bias in Claude
The post sets out an effort to make Claude politically even-handed: balanced, accurate, comprehensive, respectful, and reluctant to offer unsolicited political opinions or persuasion. It describes system-prompt guidance and reinforcement-learning-based character training, including traits that encourage neutral terminology, representation of competing perspectives, and restraint on sensitive political topics. For measurement, the Paired Prompts method compares responses to the same contentious issue framed from opposing ideological positions, scoring dimensions such as equal engagement and acknowledgment of opposing views; an automated version applies it to thousands of prompts across hundreds of stances. In the reported comparison, Claude Sonnet 4.5 was more even-handed than GPT-5 and Llama 4, and similar to Grok 4 and Gemini 2.5 Pro. The evaluation is open-sourced, with limitations including a focus on current US discourse and single-turn interactions.
Maryland partners with Anthropic to serve residents
Maryland has announced a partnership with Anthropic to deploy Claude across state agencies, aiming to improve government operations and serve more than six million residents. The partnership includes a Claude-powered virtual assistant for SNAP, Medicaid, temporary cash assistance, and WIC; it will also identify other programs residents may qualify for. Caseworkers will use Claude to process and verify more than 150,000 documents manually handled each month, validate eligibility, and receive policy guidance for complex cases. Maryland will explore an AI upskilling pilot for early-career professionals in “lighthouse” industries, while a Governor’s Innovation Team tool will use Claude to identify unmet local needs such as fresh food and childcare. The effort builds on a bilingual chatbot launched in June for more than 600,000 SUN Bucks recipients, which simplified information access and reduced call center volume.
Microsoft, NVIDIA, and Anthropic partnerships
Microsoft, NVIDIA, and Anthropic announced strategic partnerships to scale Anthropic’s Claude models on Microsoft Azure and broaden enterprise access to them. Anthropic committed to purchase $30 billion of Azure compute capacity and contract additional capacity up to one gigawatt, initially using NVIDIA Grace Blackwell and Vera Rubin systems. NVIDIA and Anthropic will collaborate on design and engineering to optimize Anthropic models for performance, efficiency, and total cost of ownership, while adapting future NVIDIA architectures to Anthropic workloads. Microsoft Foundry customers will be able to access Claude Sonnet 4.5, Claude Opus 4.1, and Claude Haiku 4.5, while Claude remains available across Microsoft’s Copilot products. Microsoft and NVIDIA committed up to $5 billion and $10 billion, respectively, to Anthropic; Amazon remains its primary cloud provider and training partner.
Claude in Microsoft Foundry and 365 Copilot
Anthropic and Microsoft are expanding their partnership by making Claude Sonnet 4.5, Haiku 4.5, and Opus 4.1 available in public preview through Microsoft Foundry, where Azure customers can build applications and enterprise agents. The integration uses serverless deployment, supports Foundry APIs, tools, and workflows, and works with existing Azure agreements, including MACC, while offering Python, TypeScript, and C# SDKs with Microsoft Entra authentication. Claude models are also available through Claude Code in Foundry, and Microsoft 365 Copilot integrations include Researcher, Copilot Studio custom agents, and a preview Agent Mode option in Excel. In Excel, Claude can generate formulas, analyze data, identify errors, and iterate on solutions. Foundry's Global Standard deployment uses standard API pricing; US DataZone availability is described as coming soon, and the models support capabilities including code execution, web search, citations, vision, tool use, and prompt caching.
Snowflake and Anthropic announce $200M partnership
Anthropic and Snowflake announced a multi-year, $200 million partnership that brings Claude models to Snowflake’s more than 12,600 customers across Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Azure. The agreement also includes a joint go-to-market initiative for deploying AI agents in large enterprises. Claude can analyze structured and unstructured data through Snowflake Cortex AI, while Snowflake Intelligence and Cortex Agents support natural-language questions, multimodal SQL-based analysis, and custom multi-agent solutions under Snowflake Horizon Catalog governance. Snowflake’s internal benchmarks report greater than 90% accuracy on complex text-to-SQL tasks, and examples include Simon Data and Intercom. The partnership is positioned to help regulated industries move AI agents from pilots to production within governed data environments.
Anthropic acquires Bun as Claude Code hits $1B
Anthropic announced the acquisition of Bun, a JavaScript runtime, as Claude Code reached $1 billion in run-rate revenue six months after becoming publicly available. Bun is an all-in-one toolkit that combines a runtime, package manager, bundler, and test runner, and the post says its focus on speed, reliability, and developer experience has made it important to Claude Code’s infrastructure. The companies had collaborated for months, with Bun helping scale Claude Code and contributing to its native installer; Anthropic says the acquisition will bring faster performance, improved stability, and new capabilities. Bun will remain open source and MIT-licensed, while Anthropic continues investing in its JavaScript and TypeScript tooling, which has more than 7 million monthly downloads and over 82,000 GitHub stars.
Donating MCP to the Agentic AI Foundation
Anthropic is donating the Model Context Protocol (MCP) to the Agentic AI Foundation (AAIF), a directed fund under the Linux Foundation co-founded by Anthropic, Block, and OpenAI, with support from Google, Microsoft, AWS, Cloudflare, and Bloomberg. Introduced as an open standard for connecting AI applications to external systems, MCP now has more than 10,000 active public servers, adoption across major AI products, enterprise deployment support, and official SDKs with 97M+ monthly downloads across Python and TypeScript. The donation places MCP alongside Block’s goose and OpenAI’s AGENTS.md as AAIF founding projects, while its existing governance model remains unchanged. MCP also has a community-driven Registry, and the November 25th specification release added asynchronous operations, statelessness, server identity, and official extensions.
Accenture and Anthropic launch partnership
Anthropic and Accenture announced an expanded partnership aimed at helping enterprises move AI projects from pilots to full-scale deployment. The companies are forming the Accenture Anthropic Business Group, a dedicated Claude-focused practice, and plan to train approximately 30,000 Accenture professionals, including engineers who embed Claude in client environments. Their first joint offering places Claude Code at the center of software development and combines productivity and ROI measurement, workflow redesign, and change management and training for CIOs. Initial industry solutions target financial services, life sciences, healthcare, and the public sector, where security and governance requirements are strict; the partnership also brings Claude to Accenture Innovation Hubs and establishes a Claude Center of Excellence for co-designing offerings.
Protecting the wellbeing of our users
Claude’s safeguards for conversations involving emotional support focus on suicide and self-harm, sycophancy, and an 18+ access requirement. For suicide and self-harm, the approach combines system-prompt guidance, reinforcement learning informed by human preference data and expert input, and a classifier that can display country-specific crisis resources supplied by ThroughLine. The post reports that Claude Opus 4.5, Sonnet 4.5, and Haiku 4.5 respond appropriately to clear-risk single-turn prompts 98.6%, 98.7%, and 99.3% of the time, while multi-turn results are 86% and 78% for Opus 4.5 and Sonnet 4.5. Prefilled real-conversation stress tests produced 91% and 73% for those models, versus 36% for Opus 4.1. The work also describes efforts to reduce sycophancy, evaluate model behavior, identify underage users, and continue publishing methods and results.
Working with the US Department of Energy
Anthropic and the US Department of Energy (DOE) announce a multi-year partnership under the Genesis Mission, targeting American energy dominance, biological and life sciences, and scientific productivity across potentially all 17 national laboratories. Anthropic plans to give DOE researchers access to Claude and engineers who can build purpose-built AI agents, Model Context Protocol servers connecting Claude with scientific instruments and tools, and Claude Skills for specialized workflows. Proposed applications include speeding permitting reviews, supporting nuclear technology research and energy security, developing pandemic and biological-threat early-warning systems, accelerating drug discovery, and using fifty years of DOE research to surface ideas and patterns. The partnership extends prior DOE work, including a nuclear risk classifier co-developed with the National Nuclear Security Administration and Claude’s rollout at Lawrence Livermore National Laboratory. Anthropic says future arrangements could develop a model for collaboration between AI and human researchers.
Topics they write about
All topics →Weekly digest
The best engineering reads, once a week.