AI News Roundup: Big Tech Bets $725B on AI, Google Pours $40B Into Anthropic, Novo Nordisk Teams With OpenAI

Big Tech’s AI Spending Hits $725 Billion

Q1 2026 earnings season delivered a jaw-dropping number: Google, Amazon, Microsoft, and Meta now plan to spend a combined $725 billion on capital expenditures this year, up 77% from last year’s record $410 billion. Nearly all of the increase is flowing into AI infrastructure — data centers, custom chips, and networking gear.

Amazon leads the pack with a $200 billion budget, followed by Microsoft at $190 billion, Alphabet at $180–$190 billion, and Meta at $125–$145 billion (raised from its prior guidance of $115–$135 billion). Rising memory chip prices pushed forecasts higher at both Microsoft and Meta, with component cost inflation accounting for roughly $25 billion of Microsoft’s total. Only Google convinced investors the spending is paying off — Google Cloud revenue grew 63% year-over-year to $20 billion, sending Alphabet stock to its best monthly gain since 2004.

Google Invests Up to $40 Billion in Anthropic

Google announced it will invest up to $40 billion in Anthropic — $10 billion now at a $350 billion valuation, with up to $30 billion more contingent on performance milestones. The deal cements Google’s dual role as both Anthropic’s competitor and its largest infrastructure supplier, providing access to Google’s tensor processing units and cloud platform.

The investment comes alongside a broader compute partnership. Anthropic, Google, and chipmaker Broadcom announced a deal to build 5 gigawatts of TPU-based computing capacity, set to come online in 2027. Combined with a separate $100 billion, 5-gigawatt compute agreement with Amazon, Anthropic is locking in an unprecedented scale of infrastructure for training and running future Claude models.

Anthropic Overtakes OpenAI in LLM Revenue

New data from Counterpoint Research puts Anthropic at the top of the global LLM revenue table for Q1 2026, with a 31.4% market share versus OpenAI’s 29%. Anthropic’s annualized run-rate revenue has surged past $30 billion, up from roughly $9 billion at the end of 2025.

The research firm attributes Anthropic’s lead to its grip on the high-end professional market, where customers pay for output quality rather than novelty. Claude Opus 4.6 and GPT-5.5 remain neck-and-neck on benchmarks — Opus leads on 6 of 10 shared evaluations — but Anthropic’s enterprise pricing and reliability appear to be winning the revenue race.

Novo Nordisk Partners With OpenAI for Drug Discovery

Danish pharmaceutical giant Novo Nordisk announced a strategic partnership with OpenAI to deploy AI across its entire business — from drug discovery and clinical trials to manufacturing and supply chains. Pilot programs will launch across R&D, manufacturing, and commercial operations, with full integration targeted by the end of 2026.

The timing is no accident: Novo Nordisk is locked in a fierce battle with Eli Lilly for dominance in the weight-loss drug market, where it has lost its first-mover advantage. The company is banking on AI to accelerate its pipeline of next-generation treatments while OpenAI gets a marquee pharma customer to prove out enterprise AI in healthcare.

EU AI Act Talks Collapse After 12-Hour Marathon

EU member states and Parliament failed to reach a deal on proposed changes to the AI Act after 12 hours of negotiations, pushing talks into May. The sticking point: the AI Omnibus package, which aims to postpone compliance deadlines to December 2027 for standalone high-risk systems and August 2028 for those embedded in regulated products.

The delay reflects growing tension between Europe’s desire to regulate AI and fears that strict rules will push companies to build elsewhere. Separately, the European Commission is considering classifying ChatGPT under the Digital Services Act’s strict online platform rules — a move that would subject OpenAI to some of the bloc’s toughest digital regulations.

MCP Crosses 97 Million Installs Under Linux Foundation Governance

Anthropic’s Model Context Protocol hit 97 million monthly SDK downloads in March 2026, up from 2 million at launch in November 2024. Every major AI provider now ships MCP-compatible tooling, and the protocol is governed by the Linux Foundation’s Agentic AI Foundation, co-founded by Anthropic, Block, and OpenAI.

The move to open governance has eliminated the single-vendor risk that kept enterprise architects cautious, putting MCP on the same stability trajectory as Kubernetes and PyTorch. The 2026 roadmap focuses on making Streamable HTTP work at scale — fixing stateful session issues with load balancers and adding standard service discovery.