AI News for February 25, 2026 — 7min.ai

[NEWS] Anthropic drops flagship safety pledge, says it won't halt AI development if competitors 'blaze ahead': Anthropic scrapped the central commitment of its Responsible Scaling Policy: the 2023 promise to never train an AI system without guaranteeing adequate safety measures first.
[NEWS] Hegseth gives Anthropic Friday deadline to accept Pentagon's 'any lawful use' terms or face blacklisting: Defense Secretary Pete Hegseth set a Friday deadline for Anthropic to agree to the Pentagon's demand for "any lawful use" terms for Claude, which would permit mass surveillance and lethal autonomous weapons.
[TOOLS] Anthropic launches Claude Cowork with enterprise plugin marketplace for Excel, PowerPoint, and finance tools: Anthropic rolled out a sweeping update to Claude Cowork, its AI productivity platform for knowledge workers, adding a private plugin marketplace, prebuilt templates for HR, finance, engineering, and design, plus native integrations with Excel and PowerPoint.
[NEWS] Meta signs multi-year deal with AMD for up to 6GW of AI inference chips, buys 10% equity stake: Meta and AMD announced a multi-year deal covering up to six gigawatts of Instinct GPUs focused on inference workloads.
[NEWS] Wayve raises $1.5B at $8.6B valuation to take its robotaxi software global: UK-based autonomous vehicle startup Wayve raised $1.5B from Microsoft, Nvidia, Uber, Mercedes-Benz, Nissan, and Stellantis to expand its self-driving software licensing business.
[NEWS] AI chip startup MatX raises $500M Series B to challenge Nvidia with 10x LLM training performance: MatX, founded by two former Google TPU engineers, raised a $500M Series B led by Jane Street and Situational Awareness, the investment fund of former OpenAI researcher Leopold Aschenbrenner.
[TOOLS] Anthropic launches Claude Code Remote Control, bringing mobile access to coding agents: Anthropic released Remote Control, a new mode that lets developers issue commands to Claude Code from iPhone and Android devices.
[NEWS] Amazon AGI lab head David Luan departs the company: David Luan, who led Amazon's artificial general intelligence lab and its web agents effort, announced he will leave the company at the end of the week.
[RESEARCH] Inception Labs launches Mercury 2, first diffusion-based reasoning LLM hitting 1,009 tokens/sec: AI startup Inception Labs released Mercury 2, a reasoning model that replaces standard autoregressive decoding with a diffusion-based approach, generating multiple tokens simultaneously.
[NEWS] Moody's flags $662B in hidden data center liabilities across five hyperscalers: A Moody's Ratings report found the top five US hyperscalers (Amazon, Meta, Alphabet, Microsoft, Oracle) have accumulated $662B in future data center lease commitments that sit entirely off their balance sheets.

About 7min.ai · Subscribe to newsletter