Two dozen companies and organisations have come together to urge US policymakers to protect open-weight artificial intelligence models. The letter, published today in a downloadable PDF format, carries signatures from a diverse range of direct commercial rivals as well as those with little obvious overlap in their business model. These include Meta, Microsoft, Nvidia, IBM, Dell Technologies, CrowdStrike, Palantir, ServiceNow, Hugging Face, Perplexity, and Mistral.
The letter's authors, who claim to represent a wide array of industries, including cybersecurity and data analytics, argue that open-weight AI models are not only increasingly sophisticated but also pose significant risks. They contend that these models have the potential to be used for malicious purposes, such as hacking and personal surveillance, and that policymakers must take action to safeguard them.
The signatories stress that they come from a variety of sectors, including direct commercial rivals and organisations with little obvious overlap in their business model. This suggests that the letter is not driven by ideological or partisan interests but rather by a shared concern for the potential risks and consequences of open-weight AI models. As one analyst noted, this could mark an important shift in the tech industry's approach to artificial intelligence, with companies from different sectors coming together to demand greater protection for these powerful technologies.