Anthropic Names National Security Expert to Its Governing Trust

Just after unveiling new AI models for U.S. defense use, Anthropic named Richard Fontaine—ex-adviser to Sen. McCain—to its long-term benefit trust. The trust helps steer the company’s ethical path, prioritizing safety over profits. CEO Dario Amodei says Fontaine’s security expertise comes at a key moment as AI and defense increasingly overlap. Anthropic, like OpenAI and Google, is eyeing deeper national security ties.

Read more on: TechCrunch

Windsurf reports Anthropic is restricting direct access to Claude AI models

Anthropic has reduced Windsurf’s access to its Claude 3.5 and 3.7 models, prompting CEO Varun Mohan to express disappointment over the short notice. Windsurf, reportedly being acquired by OpenAI, now relies on third-party providers and user-supplied API keys. This limits access to Claude 4, frustrating users and complicating competition in the fast-growing vibe coding space.

Read more on: TechCrunch

Claude Launches Voice Mode for Hands-Free AI Conversations

Claude’s mobile app now features Voice Mode, allowing users to speak and listen to AI responses. Rolling out in English, it offers voice customization, real-time transcription, and Google Workspace integration for paid users. Ideal for multitasking, brainstorming, and planning, Voice Mode supports seamless switching between text and voice while maintaining chat context.

Read more on:Anthropic

Anthropic Levels Up with Claude 4 — But Is It Enough to Shake the AI Giants?

At its developer conference, Anthropic made a bold statement with the debut of two new models: Claude Opus 4 and Claude Sonnet 4. More than just technical upgrades, this launch signals Anthropic’s deepening ambition to become a major force in the AI landscape — even as rivals like OpenAI and Google keep pushing the limits.

Source: TechCrunch

Apple Teams Up with Anthropic for AI-Powered Coding Tool

Apple is reportedly working with Amazon-backed startup Anthropic to develop an AI-powered “vibe coding” platform that can write, edit, and test code using the Claude Sonnet model. The tool, based on Xcode, is currently for internal use, with no public release confirmed. This move underscores Apple’s growing AI ambitions amid industry-wide competition.

Read more on finance.yahoo

Anthropic Boosts Claude with Google Workspace Integration and Smart Research

Anthropic has upgraded Claude with two big features: direct access to Gmail, Calendar, and Docs, turning it into a true workplace assistant, and a new “Research” tool that runs multi-step searches with citations. These upgrades aim to rival Microsoft’s Copilot, while Anthropic promises more powerful updates soon — all while addressing privacy and accuracy concerns.

Read more on TheVerge

Anthropic CEO Warns Spies Target $100M AI Breakthroughs in Mere Code Snippets

Anthropic’s CEO, Dario Amodei, is sounding the alarm about Chinese spies possibly nabbing valuable AI secrets—think $100 million ideas in just a few code lines. Speaking at a recent event, he urged the U.S. government to boost security for AI labs, citing China’s espionage reputation and potential misuse of AI tech.

Read more on TechCrunch

Anthropic’s Claude 3.7 Sonnet Brings Smarter AI Thinking

Anthropic’s new Claude 3.7 Sonnet is here, a “hybrid” AI that can answer fast or think deeply—your choice! It’s pricier than some rivals but shines in real-world coding (62.3% on SWE-Bench). Free users get a basic version, while premium folks unlock reasoning perks. Plus, a cool coding tool, Claude Code, is in preview!

Read more on TechCrunch

X