Big Tech Unites for Open Source AI—and Against Anthropic

What It Covers

A broad coalition of big tech leaders—including Jensen Huang, Sundar Pichai, Satya Nadella, and Mark Zuckerberg—signs an open letter defending open-weight AI models against potential US government restrictions, while Anthropic stands alone in opposition, and NVIDIA backs Ilya Sutskever's secretive Safe Superintelligence lab with compute resources.

Key Questions Answered

• •Open Source Coalition Dynamics: When evaluating industry consensus, note who is absent. Every major tech company—Microsoft, Google, Meta, NVIDIA, even OpenAI—signed the open-weight AI letter, but Anthropic refused. Anthropic CEO Dario Amodei has publicly stated that widely available models with Mythos-class cyber capabilities represent a serious, concrete risk worth resisting even under significant industry pressure.
• •China Chip Independence Trajectory: China reduced US chip dependency from 90% in 2021 to 60% by 2025, with Morgan Stanley projecting a further drop to 25% by 2030. Huawei alone plans to ship 1.5 million AI chips in 2025—double its prior year total. A Shanghai firm has begun mass-producing deep ultraviolet lithography machines previously blocked by export controls.
• •Distillation vs. IP Theft Distinction: The open letter draws a deliberate line between model distillation—using one model's outputs to train another, a standard industry practice—and unlawful extraction of proprietary model weights. Policymakers and builders should track this distinction closely, as US policy responses targeting distillation broadly could eliminate a foundational technique used across the entire AI development ecosystem.
• •SSI Compute Scale Signal: NVIDIA's undisclosed investment in Safe Superintelligence includes access to next-generation Vera Rubin chips, enabling SSI to 10x its compute resources within 12 months. SSI previously relied on Google TPUs. The deal signals SSI has reached a research milestone worth scaling—a rare public signal from a lab that has operated in near-total secrecy since its 2023 founding.
• •Regulatory Capture Risk for Builders: The open letter explicitly frames open-weight AI as a defense against market centralization, not just a technical preference. For builders and enterprise leaders, this matters practically: a US ban on Chinese open-source models could trigger a mass extinction event for startups currently building on those model foundations, as argued by Y Combinator and the Little Tech Association.