Anthropic has come under criticism for being the only major frontier AI company that did not sign an industry letter supporting open-weight AI models. While companies including Nvidia, Microsoft, Meta, Google, OpenAI, and SpaceX endorsed the initiative, Anthropic chose not to join, drawing criticism from investors, industry leaders, and some competitors who argue that open-weight models are important for innovation, competition, and U.S. AI leadership.
The debate centers on whether advanced AI model weights should be made publicly available for developers to download, modify, and deploy. Anthropic has consistently favored a more controlled approach, with CEO Dario Amodei arguing that unrestricted access to highly capable AI models could increase security risks and make it easier for malicious actors to misuse the technology. Critics, however, contend that the company's position also helps protect its commercial advantage by limiting competition from open-weight alternatives.
The controversy has intensified as Chinese AI companies release increasingly capable open-weight models, including Kimi K3, raising concerns about global AI competitiveness. At the same time, U.S. officials are investigating allegations that some Chinese models may have been trained using distilled outputs from proprietary Western AI systems, prompting discussions about sanctions and stronger intellectual property protections.
The disagreement highlights a broader divide within the AI industry over the future of model development. Supporters of open-weight AI believe wider access accelerates research, transparency, and innovation, while companies like Anthropic argue that frontier AI requires tighter safeguards to reduce the risk of misuse. As governments consider new AI regulations, the balance between openness, security, and commercial competition is expected to remain one of the industry's most significant policy debates.
