A coalition of 25 companies and organizations — spanning chipmakers, cloud providers and cybersecurity firms — has signed a joint letter arguing that open-weight AI models are essential to keeping the U.S. at the front of the AI race.

Which Companies Signed

The letter, titled “Open Weights and American AI Leadership” and dated July 24, 2026, carries signatures from: 

  • NVIDIA Corp. (NASDAQ:NVDA)
  • Meta Platforms Inc (NASDAQ:META)
  • Microsoft Corp. (NASDAQ:MSFT) 
  • Palantir Technologies (NASDAQ:PLTR)
  • CrowdStrike Holdings (NASDAQ:CRWD) 
  • IBM (NYSE:IBM)
  • ServiceNow (NYSE:NOW)
  • Hugging Face
  • Mistral AI
  • Mozilla 
  • Linux Foundation
  • Andreessen Horowitz
  • Y Combinator
  • Reflection
  • Replit
  • Telnyx
  • Arcee AI
  • Arena
  • Black Forest Labs
  • Emergence Capital
  • Mariana Minerals
  • The American Innovators Network
  • Perplexity

An Open-Source Rerun

The letter opens by invoking the 1980s open-source software movement, which pushed back against the idea that software could only advance behind closed doors. The movement, the signers note, now underpins most of the internet and critical U.S. military and federal systems. 

They argue AI is facing the same fork in the road — and that open-weight models, which anyone can download, inspect, modify and run, need to be part of America’s foundation.

Read Also: Jensen Huang Posts On X For The First Time Ever — And Uses It To Defend Open-Source AI

The Economic Case

Open weights, the letter argues, allow startups, enterprises, universities and public institutions to build on advanced AI without training a model from scratch or paying frontier-model prices for every task — matching the right model to the right job at the right cost. 

That efficiency, signers say, is what makes AI economically sustainable as usage scales into the billions of daily tasks across factories, hospitals, classrooms and small businesses.

Competition and Control

The coalition also frames open weights as a competitive safeguard, spreading rivalry across model developers, cloud providers, chipmakers and application builders rather than letting gains concentrate among a few firms.

For enterprise customers, open weights offer protection against vendor lock-in, letting them control their own data and adapt models to their needs.

Safety, Not Just Risk

On security, the letter concedes open weights carry real risks — once released, they’re outside the original developer’s control. But it argues closed models aren’t inherently safer, since they can be breached or misused without outside detection, and that concentrating AI capability behind a few providers creates dangerous single points of failure.

Distillation Pushback

The letter also draws a line on distillation, calling it a legitimate, widely used model-improvement technique rather than “misappropriation,” and urges policymakers to address genuine IP theft through targeted legal frameworks rather than broad restrictions that could chill innovation.

Read Also: Intel’s AI Boom Raked in $2.5 Billion—Then Its Foundry Lost 84% of It

Photo: Summit Art Creations / Shutterstock