Anthropic CEO Defends AI Safety Record Ahead of IPO
Dario Amodei defended Anthropic's AI safety protocols against accusations of deprioritizing security in a race toward superintelligence.
Anthropic CEO Dario Amodei defended the company's commitment to AI safety following public criticism from Nathan Calvin, the chief strategy officer of Encode AI. In an exchange on X, Calvin accused the company of undermining its safety commitments to win a race toward superintelligence, pointing to massive compute deals and the use of the Claude model to develop new AI systems.
Amodei responded by describing a "race to the top" strategy, arguing that Anthropic can better secure the industry by remaining at the technological frontier and demonstrating how to build technology responsibly. He cited specific safety measures, including limiting sales to Chinese state-controlled firms and restricting the Mythos model to cybersecurity partners. While defending the company's track record, Amodei admitted that Anthropic could have better handled a draft bill circulated to Congress.
The dispute comes as Anthropic prepares for an initial public offering in November, with the company's valuation expected to reach $2 trillion.
