Anthropic IPO Plans Delayed Amid AI Safety Warnings
Anthropic plans to hold a massive initial public offering (IPO) in November, later than many investors had anticipated. This delay comes as leaders of the largest companies in the artificial intelligence competition call for a slowdown in the development of the technology.
According to Wall Street Journal sources, the AI giant was expected to hold a record-breaking offering in October. Some Anthropic advisors stated that waiting until November would allow the startup time to share its third-quarter financial reports, which demonstrate a strong competitive position, even after its competitor, OpenAI, launched its latest model, Astra, in September.
Those familiar with the timing noted that the decision to target a November offering was made before public warnings from former Anthropic researchers triggered debates regarding whether AI progress is moving too rapidly. They cautioned that the timing remains subject to change.
Anthropic CEO Dario Amodei was the first of three major AI executives to call for a slowdown, writing a blog post on Saturday stating that the risks posed by current advanced tools are too great to continue development at the same pace. Leaders of rival AI laboratories, including OpenAI CEO Sam Altman and Elon Musk of SpaceX, have echoed this sentiment.
Investors estimate that Anthropic will be valued at approximately US$2 trillion (equivalent to Rp35,796 trillion) and aims to raise up to US$100 billion (equivalent to Rp1,789.80 trillion) in the share offering, both figures of which would surpass the records set by SpaceX’s debut in June.
Anthropic plans to meet with potential investors in the coming days, some of whom are expected to press the company on how a potential slowdown in the launch of new AI models might impact Anthropic’s finances or valuation targets.
These meetings are intended to help the company and its bankers gauge investor interest in the upcoming share offering.
Anthropic’s advisors and investors currently argue that slowing the pace of development is unlikely to have a drastic impact on its financial prospects, given the revenue that can be generated from existing models.
Investors currently estimate that the company will achieve annual revenue of more than US$110 billion by the end of the year.
There was little discussion regarding the IPO when the company gathered several investors and partners at its headquarters on Thursday. Instead, executives, including Jared Kaplan, Benjamin Mann, and Andrej Karpathy, discussed the company’s new offerings, including the Standard Hardware Model—a product that enables AI agents to operate physical products, such as microscopes or robotic arms, according to one participant.
At a dinner hosted by Anthropic at a steakhouse in San Francisco on Thursday night, venture capital investor conversations shifted to the safety warnings championed by Amodei and the news cycle they have created.
Many Silicon Valley firms that have invested in Anthropic stated that Amodeit’s timing for pushing new safety standards across the industry is sensible, and they are not concerned that it will reduce interest in the IPO. One investor noted that once Anthropic becomes a public company, it will likely face much stricter scrutiny from shareholders, and establishing its position regarding broader security risks could help protect it if unforeseen issues arise in the future.
Meanwhile, Anthropic’s largest competitor, OpenAI, has clearly stated that they do not plan to go public until 2027 and are in the early stages of discussions for new funding that could value the company at more than US$1.2 trillion, if completed.
Last time the ChatGPT developer raised capital, they successfully raised over US$120 billion from investors. Some Anthropic investors noted that if OpenAI raises a similar amount before Anthropic goes public, it could reduce demand for the Claude developer’s listing.