ChatGPT firm files confidential S-1 as Altman sets out vision for a post-AGI economy
OpenAI has submitted a confidential S-1 registration statement to the Securities and Exchange Commission, taking a formal step toward a potential initial public offering.
Also on Monday, the firm’s CEO published an expansive vision of where the company believes artificial intelligence is headed and what it expects to do with any capital it raises.
"We recently submitted a confidential S-1," the firm said. "We expect it to leak so we're just announcing it."
However, while excitement builds ahead of the SpaceX IPO, those investors wanting to take a slice of the business behind ChatGPT may have to be patient as no listing date has been set. In fact, the statement suggested it could be some time before it goes public.
"We have not decided on timing yet; it may be a while because there are things we want to do that are likely easier as a private company," OpenAI said. "But it's a complicated set of tradeoffs and this gives us the option to go public sooner if that ends up being best."
The confidential S-1 means that the SEC review can begin without making OpenAI’s financials public.
The company has been managing a complex conversion away from its nonprofit-controlled structure toward a more conventional for-profit model, a transition that has attracted legal challenges and regulatory attention. The acknowledgment that some planned moves are easier to execute as a private company suggests that process is still unresolved.
Investor pitch?
The filing arrived alongside a lengthy essay from chief executive Sam Altman and chief scientist Jakub Pachocki that amounts to a manifesto for OpenAI's next chapter and implicitly, an investor pitch.
The document, published under the title "Built to benefit everyone: our plan," frames the company's ambitions in deliberately sweeping terms, drawing a comparison to rural electrification in the 1920s as the closest historical parallel to what AI may deliver.
The essay sets out three near-term goals: building an automated AI researcher capable of accelerating the company's own scientific work; using AI to drive broader economic growth while distributing the gains widely; and providing every person on Earth with what it calls a personal AGI.
On the research timeline, Altman and Pachocki said: "Our internal belief is that by March of 2028 we may have a significant fraction of our research being done by AI systems in tandem with our own researchers."
But following the recent call by Claude-maker Anthropic (which also recently filed an S-1)for the AI industry to consider slowing down or pausing development amid risks that the technology may soon be able to build itself, the two OpenAI executives were also explicit about the risks they see in the technology they are building.
"Entirely automating everything is not the future we want," they wrote. "It would be unfulfilling, and it would be dangerous." The essay argues that as AI systems grow more capable, human judgment becomes more rather than less important with people responsible for setting direction, applying values, and deciding what is worth pursuing.
Altman and Pachocki argued that OpenAI's goal is the opposite of monopoly. "A good AI future cannot be one where a small number of institutions control most of the capability and most of the upside," they wrote, calling instead for a future in which companies, communities, and countries can all build and benefit. They also called for an international body to coordinate frontier AI development and, where necessary, slow it down.
The combination of a confidential IPO filing and a public statement of purpose on the same day is unlikely to be coincidental.
