AI Safety Center and Others Issue Joint Statement Criticizing Anthropic's Safety Commitments
1 hour ago / Read about 0 minute
Author:小编   

The AI Safety Center, joined by prominent figures from the tech industry and academia, has signed a statement highlighting a contradiction between Anthropic's safety advocacy and its actual R&D strategy. The statement was released on the eve of Anthropic's planned initial public offering. According to the signatories, Anthropic publicly warns that AI systems with self-improvement capabilities could lead to a concentration of power and a loss of human control—yet at the same time, the company is fully pursuing the development of such technology, engaging in conduct that undermines its own safety commitments, and lobbying for regulatory policies that favor its interests. The signatories include notable figures such as Harvard University psychology professor Steven Pinker, MIT professor Max Tegmark, Hugging Face co-founder Thomas Wolf, and xAI co-founder Igor Babuschkin, among others.