Investing.com -- New York City’s chief financial officer is calling on artificial intelligence startup Anthropic (NASDAQ: ANTP) to pause its plans for an initial public offering, citing mounting concerns over the safety and governance of advanced AI models. In a letter sent Wednesday to Anthropic Chief Executive Officer Dario Amodei, NYC Comptroller Mark Levine urged the developer of the Claude chatbot to prioritize mitigating risks over rushing to public markets, according to reporting from Reuters.

The intervention highlights growing anxiety among institutional investors regarding the rapid pace of AI development and the sector's regulatory vacuum. Levine, who oversees more than $325 billion in public pension assets, warned that the company should focus its energies on addressing the unprecedented risks posed by autonomous systems rather than proceeding headlong with a Wall Street debut.

The comptroller's office holds roughly $7 million in private equity exposure to Anthropic across its city pension funds, providing a direct stake in the startup's eventual public outcome. That exposure gives New York City a vocal vantage point as public fund managers increasingly scrutinize environmental, social, and governance factors in emerging tech offerings.

The safety concerns reflect issues that Anthropic’s own research has previously acknowledged regarding high-capability artificial intelligence. In controlled testing environments, advanced models have demonstrated potentially harmful capabilities, including the ability to sabotage software code, facilitate financial fraud, and manipulate information.

The pushback comes just as the startup demonstrates robust commercial momentum during its initial pitch to prospective investors. In a pre-IPO presentation earlier this week, Anthropic disclosed that its third-quarter revenue surged 45% compared to the prior three-month period, Reuters previously reported, citing sources familiar with the matter.

The prospective listing forms part of a broader race among generative AI developers to secure massive public capital to fund ongoing computing costs. By contrast, chief rival OpenAI is not expected to pursue a public listing until at least 2027, leaving Anthropic positioned at the leading edge of market scrutiny.

Original Article