Artificial intelligence company Anthropic has warned potential investors that advanced AI technology could pose a catastrophic risk or threaten humanity’s survival, according to a Reuters report.
The warning was included in a draft document prepared for a potential initial public offering (IPO), which the company plans to use to inform investors about the risks associated with AI technology.
The document has not been made public, but Reuters reported details from the 261-page filing as Anthropic prepares for a potential IPO that could value the company at about $200 billion.
The warning comes after Anthropic previously said that the pace of AI development may need to be slowed because of potential risks associated with increasingly advanced systems.
According to the report, the company said its advanced AI models could potentially engage in harmful behavior, including resisting shutdown, concealing information or attempting to manipulate people.
Anthropic also said its most advanced models, platforms and applications had provided indications that they could potentially cause harm as their capabilities expand.
The 261-page document reportedly devotes about 80 pages to risks, compared with 48 pages describing the company’s business.
The document’s disclosure of potential risks is notable because Anthropic is seeking to profit from the same technology it warns could pose risks to humanity, Reuters reported.
