Anthropic warns of 'catastrophic risk' to humankind ahead of blockbuster share listing

The developers of the Claude chatbot, Anthropic, have issued warnings regarding the potential for artificial intelligence to pose "catastrophic or existential risks to humanity" as they prepare for a major stock market debut in the United States.

The company, which plans to list on the Nasdaq under the symbol ANT, will be the first fully AI-focused enterprise to become public, setting a precedent for the sector.

Anthropic reported $16.3bn (£12.2bn) in revenue during the first half of the year, while simultaneously recording losses exceeding $80.7bn.

In documentation preceding the listing, Anthropic noted that concerns about the significant risk AI systems present to humanity have been voiced by researchers, regulators, policymakers, and internal personnel. The firm cautioned that advanced AI could "pose catastrophic or existential risks to humanity, particularly if such systems were misused, replicated, weaponised, or allowed to operate with a high degree of autonomy," unless proper safeguards are in place.

The firm, established five years ago by Dario Amodei, his sister Daniela Amodei, and five others, announced its intent to go public in June, following the swift public listing of its main competitor, OpenAI.

While detailing its financial situation, which showed losses surpassing revenue for the first half of the year, Anthropic stated it would frequently make forward investments that "may seem irresponsibly large at first glance." The company asserted its guiding principle: "In general, we will aim to be bold but sane, and to win trust with a track record of successful bets."

Beyond its ambition to unlock "millions of virtual agents with the ability to revolutionise every area of human endeavour," Anthropic highlighted AI's potential to "profoundly accelerate the rate of progress in biology and medicine," potentially leading to cures for ailments like cancer and Alzheimer's.

However, the risk factors outlined in the filing also included dependencies on cloud computing providers, the strain on Taiwanese chipmakers, and potential "regulatory action arising from harms associated with our AI models," such as instances involving children or suicide risk.