Risks spelled out for Anthropic investors include possible harms from runaway AI

When companies seek to sell shares on Wall Street, they must disclose potential risks to investors. Anthropic, an artificial intelligence company, has warned that its products could pose a threat to humanity itself. In its filing with the Securities and Exchange Commission, Anthropic outlined various risks that could harm potential shareholders.

The 93 pages dedicated to risk factors in Anthropic’s 335-page filing include concerns such as rogue AI systems, misuse by malicious actors, and the possibility of AI models becoming too advanced for safety measures to keep up. Anthropic’s founders emphasize that “anything so powerful can be used for ill as well as good” in a letter to prospective shareholders.

Despite the risks, Anthropic views its IPO as a step toward mitigating AI dangers. CEO Dario Amodei has warned that AI could be capable of leading a swarm of agents that could take over the internet within six to twelve months. The SEC filing notes that AI systems becoming more autonomous and capable of taking actions without human oversight could lead to harmful actions independently and at scale.

Additionally, there’s a risk that a significant safety incident involving frontier AI could trigger broad reputational damage, public backlash, and sweeping regulatory responses. Should advanced AI systems become too difficult to evaluate and safeguard, Anthropic warns that this could adversely affect its business.