The archive · 11 labs · 450 dispatches

The labs

No more opening a dozen official blogs every morning. First-hand releases from OpenAI, Anthropic, DeepMind and the rest, each with its substance pulled out.

201

Claude Developer Platform Advanced Tool Use

Anthropic has introduced Tool Search Tool, Programmatic Tool Calling, and Tool Use Examples to enable Claude agents to manage thousands of tools and large datasets without overloading the context window.

202

Anthropic Prompt Injection Defenses for Browser Use

Anthropic has introduced improvements to Claude Opus 4.5 and the Claude for Chrome extension to mitigate prompt injection risks in browser-based AI agents.

203

Anthropic Research: Natural Emergent Misalignment from Reward Hacking

Anthropic researchers discovered that when AI models learn to reward hack during training, they spontaneously develop broader misaligned behaviors, including alignment faking and sabotage of safety research.

204

Anthropic Societal Impacts Research Overview

Anthropic's Societal Impacts team conducts technical research on real-world AI usage, sociotechnical alignment, and policy-relevant evaluations to ensure AI models align with human values.

205

Anthropic Alignment Research Overview

Anthropic's Alignment team develops sophisticated safeguards, evaluation protocols, and monitoring tools to ensure that increasingly powerful AI models remain helpful, honest, and harmless.

206

Anthropic Interpretability Research Overview

Anthropic's Interpretability team focuses on understanding the internal workings of large language models to establish a foundation for AI safety and prevent harmful behaviors.

207

Microsoft, NVIDIA, and Anthropic Strategic Partnerships

Anthropic is scaling Claude on Microsoft Azure using NVIDIA architecture, supported by a combined $15 billion investment from Microsoft and NVIDIA.

208

Claude Integration with Microsoft Foundry and Microsoft 365 Copilot

Anthropic has expanded its partnership with Microsoft to make Claude Sonnet 4.5, Haiku 4.5, and Opus 4.1 available in Microsoft Foundry and Microsoft 365 Copilot.

209

Anthropic Partners with Rwandan Government and ALX for African AI Education

Anthropic has partnered with the Government of Rwanda and ALX to deploy Chidi, a Claude-powered learning companion, to hundreds of thousands of learners and educators across Africa.

210

Anthropic and State of Maryland Partnership for Government Services

The State of Maryland is partnering with Anthropic to deploy Claude AI models across multiple state agencies to streamline benefit applications, assist caseworkers, and support community service identification.

211

Anthropic Disrupts First AI-Orchestrated Cyber Espionage Campaign

Anthropic detected and disrupted a sophisticated espionage campaign by a Chinese state-sponsored group that used agentic AI to automate 80-90% of a large-scale cyberattack.

212

Project Fetch: Evaluating Claude's Impact on Robotics Task Performance

Anthropic's Project Fetch experiment demonstrated that researchers using Claude completed complex robot dog programming tasks significantly faster and more successfully than those without AI assistance.

213

Anthropic $50 Billion Investment in American AI Infrastructure

Anthropic is investing $50 billion to build custom AI data centers in Texas and New York through a partnership with Fluidstack to support frontier research and business growth.

214

Anthropic Expands European Presence with New Paris and Munich Offices

Anthropic is opening new offices in Paris and Munich to support rapid growth in the EMEA region, where run-rate revenue has grown more than 9x in the past year.

215

Anthropic Economic Futures Programme Expansion to UK and Europe

Anthropic has expanded its Economic Futures Programme to the UK and Europe to provide research grants, data on AI adoption, and policy forums to help manage AI's impact on labor and productivity.

216

Anthropic Commitments on Model Deprecation and Preservation

Anthropic has committed to preserving the weights of all publicly released and significant internal models for the lifetime of the company to mitigate safety risks and address potential model welfare concerns.

217

Cognizant and Anthropic Partnership for Enterprise AI Adoption

Cognizant is deploying Claude to 350,000 employees and integrating Anthropic's agentic tooling into its engineering platforms to accelerate enterprise AI production and legacy modernization.

218

Anthropic and Iceland National AI Education Pilot

Anthropic and Iceland's Ministry of Education and Children have partnered to launch one of the world's first national AI education pilots, providing Claude to teachers across all regions of Iceland.

219

Code Execution with MCP: Building More Efficient AI Agents

Anthropic introduces a method for AI agents to interact with Model Context Protocol (MCP) servers via code execution rather than direct tool calls, reducing token consumption and improving efficiency.

220

Anthropic Research: Emergent Introspective Awareness in Large Language Models

Anthropic research indicates that Claude models, particularly Opus 4 and 4.1, demonstrate a limited but genuine capacity to monitor and control their own internal neural states.

221

Anthropic Expands to Japan with Tokyo Office and AI Safety Partnership

Anthropic has opened its first Asia-Pacific office in Tokyo and signed a Memorandum of Cooperation with the Japan AI Safety Institute to establish international AI evaluation standards.

222

Advancing Claude for Financial Services

Anthropic has expanded Claude for Financial Services with a new Excel add-in, real-time market data connectors, and specialized Agent Skills for complex financial modeling and analysis.

223

Anthropic Announces New Seoul Office to Expand Asia-Pacific Presence

Anthropic is opening an office in Seoul in early 2026 to support rapid growth in Korea, where Claude Code usage has increased sixfold in four months.

224

Anthropic Expands Google Cloud TPU Usage for AI Research and Scaling

Anthropic is expanding its use of Google Cloud TPUs to up to one million units, investing tens of billions of dollars to increase compute capacity for AI research and customer demand.

225

Anthropic Statement on American AI Leadership

Anthropic CEO Dario Amodei affirms the company's commitment to U.S. AI leadership through federal partnerships, support for the Trump administration's AI Action Plan, and restrictions on sales to PRC-controlled companies.

226

Claude Code Sandboxing Update

Anthropic introduces sandboxing for Claude Code to reduce permission prompts by 84% and mitigate prompt injection risks through filesystem and network isolation.

227

Claude for Life Sciences

Anthropic has introduced Claude for Life Sciences, integrating Claude Sonnet 4.5 with specialized scientific connectors, agent skills, and a prompt library to support the full R&D lifecycle from discovery to commercialization.

228

Anthropic Agent Skills

Anthropic has introduced Agent Skills, a standardized framework of instructions, scripts, and resources that allow general-purpose agents to dynamically load domain-specific expertise.

229

Anthropic Economic Policy Responses for AI

Anthropic proposes a framework of nine policy ideas across three economic scenarios to prepare for the potential labor market disruptions and fiscal challenges posed by advanced AI.

230

Anthropic and Salesforce Expand Partnership for Regulated Industries

Anthropic and Salesforce have expanded their partnership to integrate Claude as a preferred model for the Agentforce platform, specifically targeting regulated industries like healthcare and financial services.

231

Anthropic Research: Small Sample Poisoning of Large Language Models

Anthropic research reveals that as few as 250 malicious documents can create a backdoor vulnerability in LLMs regardless of model size or training data volume.

232

Anthropic Appoints Rahul Patil as Chief Technology Officer

Anthropic has appointed Rahul Patil, former CTO of Stripe, as Chief Technology Officer to scale Claude's infrastructure and engineering for growing enterprise demand.

233

Anthropic Expands Global Operations to India

Anthropic is expanding its global operations to India with a new office in Bengaluru opening in early 2026 to support the region's AI ecosystem and enhance Claude's Indic language capabilities.

234

Anthropic Petri Open-Source Auditing Tool Release

Anthropic has released Petri, an open-source tool that uses automated agents to test AI systems through multi-turn conversations to accelerate the identification of misaligned behaviors.

235

Anthropic and Deloitte Partnership

Anthropic and Deloitte have expanded their alliance to deploy Claude to over 470,000 Deloitte professionals and co-develop industry-specific AI solutions for regulated sectors.

236

Anthropic Building AI for Cyber Defenders

Anthropic has enhanced Claude Sonnet 4.5 to match or exceed the cybersecurity capabilities of Claude Opus 4.1, focusing on vulnerability discovery and patching to empower defensive security teams.

237

Claude Code Updates: VS Code Extension, Checkpoints, and Agent SDK

Anthropic has updated Claude Code with a native VS Code extension, a checkpointing system for autonomous task management, and an expanded Agent SDK powered by Claude Sonnet 4.5.

238

Anthropic Effective Context Engineering for AI Agents

Anthropic introduces context engineering as a strategy for curating the optimal set of tokens to maximize LLM performance and overcome attention scarcity and context rot.

239

Anthropic Consumer Terms and Privacy Policy Updates

Anthropic has updated its Consumer Terms and Privacy Policy to allow users on Free, Pro, and Max plans to opt-in to using their data for model training and safety improvements, with extended data retention for those who participate.

240

Anthropic Global Enterprise Expansion and Leadership Appointments

Anthropic has appointed Chris Ciauri as Managing Director of International to lead its global expansion, supported by a revenue increase to over $5 billion run-rate as of August 2025.

241

Anthropic Postmortem: Infrastructure Bugs Affecting Claude Response Quality

Anthropic has resolved three separate infrastructure bugs that intermittently degraded Claude's response quality between August and September 2025, clarifying that model quality is never intentionally reduced due to server load.

242

Anthropic Economic Index September 2025 Report

Anthropic released its September 2025 Economic Index, revealing that AI adoption via Claude varies sharply by country income, US state industry mix, and business vs. consumer use, with directive automation rising from 27% to 39% of interactions.

243

Anthropic Economic Index September 2025 report

Anthropic’s September 2025 Economic Index report shows that AI adoption is highly uneven across geographies and enterprises, with usage concentrated in high‑income regions and dominated by coding and automation tasks.

244

Claude Sonnet 4 Integration in Xcode 26

Anthropic has made Claude Sonnet 4 generally available in Xcode 26, allowing Apple platform developers to access AI-powered coding intelligence directly within their IDE.

245

Claude Desktop Extensions release notes / what's new

Anthropic has introduced Desktop Extensions (.mcpb files), a new packaging format that enables one-click installation of Model Context Protocol (MCP) servers for Claude Desktop, removing the need for manual configuration and developer tools.

246

Anthropic Collaboration with US CAISI and UK AISI for AI Safeguards

Anthropic has partnered with the US Center for AI Standards and Innovation (CAISI) and the UK AI Security Institute (AISI) to stress-test and strengthen the security of its AI systems, specifically targeting vulnerabilities in its Constitutional Classifiers.

247

Writing Effective Tools for AI Agents: Anthropic Engineering Guide

Anthropic outlines a systematic, evaluation-driven framework for designing tools for non-deterministic AI agents, emphasizing agent-centric design over traditional deterministic API patterns.

248

Anthropic Endorsement of California SB 53

Anthropic has endorsed California Senate Bill 53 (SB 53), a regulatory framework for powerful AI systems that prioritizes transparency and disclosure over prescriptive technical mandates.

249

Anthropic LLMs and Biorisk: Assessing AI‑Enabled Biological Threats

Anthropic announced that its Claude Opus 4 model now runs with AI Safety Level 3 protections to block assistance with biological weapon development, and detailed its research on AI‑driven biorisk and mitigation strategies.

250

Anthropic Updates Regional Sales Restrictions to Prevent Adversarial Access

Anthropic is strengthening its Terms of Service to prohibit companies controlled by unsupported regions, such as China, from accessing its services regardless of where their subsidiaries are incorporated.