Nvidia CEO Jensen Huang has publicly backed an open letter defending open-weights AI models, pushing back against growing scepticism from firms like Anthropic and OpenAI. The intervention matters because Nvidia's business depends on a broad, competitive AI ecosystem; if access to open models narrows, particularly Chinese ones, American buyers would be left choosing only between mediocre domestic options or increasingly costly proprietary systems, shrinking Nvidia's market and growth prospects.

The letter, published on Friday, was signed by Meta, Microsoft, IBM and Dell alongside Nvidia, arguing that open models bolster safety, cybersecurity, innovation and national sovereignty. Analysts note the US lags China in open-weights development, with only Thinking Machine Labs' Inkling model as a credible domestic alternative to offerings from DeepSeek, MiniMax, Z.AI and Moonshot's Kimi. Huang also holds significant leverage: Nvidia has pledged billions in investment to OpenAI ($30bn) and Anthropic ($10bn), and reportedly may commit up to $250bn to back OpenAI's datacentre lease with SoftBank, giving it strong influence to nudge both firms back toward openness.