Pro-AI PACs clash over NY race tied to AI safety disclosure bill
TechCrunch · February 20, 2026
Two pro-AI super PACs are dueling over New York congressional candidate Alex Bores, with an Anthropic-funded group backing him while a rival PAC attacks. Bores is behind the RAISE Act, which would require AI developers to disclose safety protocols and report serious system misuse. The fight highlights how AI policy is becoming a direct political battleground, with industry-aligned funding shaping the debate over transparency and accountability.
Read full article →Peak XV raises $1.3B to ramp AI bets as India VC competition intensifies
TechCrunch · February 20, 2026
Peak XV has raised $1.3B in new capital, with most of it earmarked for India as the firm prioritizes AI, fintech, and cross-border investments. The raise signals sustained conviction in India’s AI-driven startup pipeline even as global VC firms crowd the market and Peak XV navigates recent partner departures.
Read full article →General Catalyst pledges $5B for India investments over five years
TechCrunch · February 20, 2026
General Catalyst says it will commit $5B to India over the next five years, a major increase from its prior $500M–$1B earmark. The move signals rising conviction in India’s startup and AI ecosystem and could intensify competition for top deals while accelerating late-stage funding and expansion capital in the region.
Read full article →Anthropic alleges Claude distillation by Chinese labs amid US chip talks
TechCrunch · February 23, 2026
Anthropic says DeepSeek, Moonshot, and MiniMax used roughly 24,000 fake accounts to mine Claude outputs and distill its capabilities into their own models. The claim lands as U.S. officials weigh tighter AI chip export controls, highlighting how model theft and hardware policy are becoming intertwined levers in the US–China AI race.
Read full article →Pentagon presses Anthropic over Claude’s military use
TechCrunch · February 23, 2026
Defense Secretary Pete Hegseth has called Anthropic CEO Dario Amodei to the Pentagon to address concerns about the military’s use of Claude, reportedly warning he could label Anthropic a “supply chain risk.” The move signals rising government pressure on frontier AI vendors, where access, controls, and procurement leverage are becoming central to how models are deployed in defense settings.
Read full article →OpenAI VCs also back Anthropic, eroding traditional investor loyalty
TechCrunch · February 23, 2026
TechCrunch reports that at least a dozen VCs who invested in OpenAI have also backed Anthropic, blurring the old norm that investors avoid funding direct competitors. Some overlaps are expected, but the scale suggests conflict-of-interest concerns are being deprioritized as AI becomes a must-own category. The shift could reshape how startups choose investors, how sensitive information is handled, and how governance expectations evolve in top-tier AI deals.
Read full article →Pentagon Presses Anthropic to Loosen AI Guardrails by Friday
TechCrunch · February 24, 2026
The Pentagon has issued Anthropic a deadline to relax certain AI safety restrictions or risk penalties, escalating a dispute over how its models can be used in defense contexts. The standoff highlights the growing tension between government procurement demands and vendor-imposed safety policies—and raises broader questions about lock-in, leverage, and investor confidence in defense-oriented AI suppliers.
Read full article →Meta eyes up to $100B AMD AI chip deal to scale beyond Nvidia
TechCrunch · February 24, 2026
Meta is reportedly entering a multiyear deal to buy billions of dollars of AMD AI chips, with pricing tied to a 160M-share warrant. The move signals a major push to diversify its AI hardware stack beyond Nvidia while accelerating data center expansion—important both for Meta’s model roadmap and for AMD’s growing role in hyperscaler AI infrastructure.
Read full article →Anthropic targets enterprises with agent plug-ins for key workflows
TechCrunch · February 24, 2026
Anthropic is rolling out enterprise-focused agent plug-ins aimed at finance, engineering, and design workflows. The move is designed to broaden Anthropic’s enterprise footprint while putting pressure on SaaS tools that currently own these task-specific functions, signaling a shift toward AI agents as a new integration layer in business software.
Read full article →Anthropic buys Vercept to boost its computer-use AI agents
TechCrunch · February 25, 2026
Anthropic has acquired Seattle-based agent startup Vercept, which built “computer-use” agents that can operate inside apps like a human user. The deal follows Meta hiring away one of Vercept’s founders, underscoring intensifying competition for agent talent and tooling. For Anthropic, it’s a fast path to stronger end-to-end automation capabilities that could expand what Claude-powered agents can reliably do on real desktops.
Read full article →Nvidia posts another record quarter as AI capex hits new highs
TechCrunch · February 25, 2026
Nvidia reported another record quarter, with CEO Jensen Huang pointing to “exponential” growth in token demand as the key driver. The results underscore how hyperscalers and enterprises are sustaining record AI infrastructure capex, reinforcing Nvidia’s central role in the buildout while signaling continued pressure on compute supply and pricing across the stack.
Read full article →Backlash Builds as AI Data Centers Face Bans and Tougher Rules
TechCrunch · February 25, 2026
Public opposition to the data-center boom is escalating, with some communities pushing for aggressive measures including outright bans on new construction. For AI companies and cloud providers, this raises the risk of slower capacity expansion, higher costs, and more complex siting and permitting—potentially constraining compute growth just as demand keeps accelerating.
Read full article →Anthropic Rejects Pentagon Bid for Unrestricted AI Access
TechCrunch · February 26, 2026
Anthropic CEO Dario Amodei says the company won’t grant the Pentagon unrestricted access to its AI systems as a deadline approaches. The standoff highlights growing tension between national security demands and AI labs’ control, safety policies, and liability exposure—and could set a precedent for how frontier model access is negotiated with governments.
Read full article →Google unveils Nano Banana 2, its latest AI image generator
Hacker News · February 26, 2026
Google has published details on Nano Banana 2, a new iteration of its AI image generation model, positioning it as an upgrade in quality and capability over prior versions. The release matters because it signals continued rapid iteration in foundation image models and raises fresh questions about access, safety guardrails, and how these tools will be integrated into Google’s broader product stack.
Read full article →Mistral AI teams up with Accenture as enterprise AI demand grows
TechCrunch · February 26, 2026
Mistral AI has struck a partnership with Accenture, adding a major global systems integrator to help deploy its models for enterprise clients. Accenture has also recently partnered with OpenAI and Anthropic, signaling a multi-vendor strategy as customers push for choice, governance, and region-specific options. For Mistral, it’s a path to scale distribution and production deployments beyond direct sales.
Read full article →