OpenAI could also be constructing as much as one of many largest preliminary public choices ever, however CEO Sam Altman says he isn’t essentially trying ahead to helming a public firm.
“Am I excited to be a public firm CEO? 0%,” Altman mentioned in an episode of the “Huge Know-how Podcast” printed on Thursday. “Am I excited for OpenAI to be a public firm? In some methods, I’m, and in some methods I believe it’d be actually annoying.”
OpenAI is laying the groundwork for an IPO, with a Thursday report from The Wall Avenue Journal placing early talks of a valuation at $830 billion. In a extra lofty estimate, the corporate may very well be valued at as much as $1 trillion, Reuters reported in October, citing three sources. In line with the Reuters report, chief monetary officer Sarah Friar is eyeing a 2027 itemizing, with a possible IPO submitting in late 2026.
Altman advised “Huge Know-how” he didn’t know if his AI firm would go public subsequent 12 months and was mum on particulars about fundraising, or the corporate’s valuation. OpenAI didn’t reply to Fortune’s request for remark.
Regardless of his hesitance to steer a public firm—which are sometimes beneath extra scrutiny, higher regulatory oversight, and are related to much less affect from founders—OpenAI’s IPO wouldn’t be all unhealthy, Altman famous.
“I do suppose it’s cool that public markets get to take part in worth creation,” he mentioned. “And in some sense, we will probably be very late to go public in the event you take a look at any earlier firm. It’s fantastic to be a non-public firm. We’d like a lot of capital. We’re going to cross the entire shareholder limits and stuff sooner or later.”
An IPO would pave the way in which for OpenAI to lift the billions of {dollars} wanted to compete within the AI race. Based as a nonprofit in 2015, OpenAI simply accomplished a fancy restructuring in October that transformed it right into a extra conventional for-profit firm, giving the nonprofit controlling the corporate a $130 billion stake in it. The restructuring additionally gave Microsoft a decreased 27% stake within the firm, in addition to elevated analysis entry, whereas concurrently liberating up OpenAI to make offers with different cloud-computing companions.
Extra ‘code reds’ to return
OpenAI’s urgency to compete with rivals was obvious earlier this month when Altman declared a “code pink” in an inside memo, following the surge of curiosity after Google rolled out its new Gemini 3 mannequin in simply someday, which the corporate mentioned was the quickest deployment of a mannequin into Google Search. Altman’s “code pink” was an eight-week mandate to redouble OpenAI’s personal efforts whereas quickly suspending different initiatives, corresponding to promoting and increasing e-commerce choices.
The blitz seems to be paying off: Final week, OpenAI launched its new GPT-5.2 mannequin, and earlier this week, it launched a brand new image-generation mannequin to compete with Google’s Nano Banana. Fidji Simo, OpenAI’s CEO of functions, mentioned the replace wasn’t in response to Google’s Gemini 3, however that the additional assets from the code pink did assist expedite its debut.
As OpenAI tries to deal with slowing person progress and retain and develop market share from its rivals, Altman conceded a code pink won’t be a one-off phenomenon. The all-out effort is a mannequin that’s been employed by Google, and likewise Meta via Fb’s extra excessive “lockdown” durations. He downplayed the stakes of a code pink, matching what sources advised Fortune equated to a centered, however not panicked, workplace surroundings.
“I believe that it’s good to be paranoid and act rapidly when a possible aggressive risk emerges,” Altman mentioned. “This occurred to us previously. That occurred earlier this 12 months with DeepSeek. And there was a code pink again then, too.”
Altman likened the urgency of a code pink to the start of a pandemic, the place motion taken in the beginning, extra so than actions taken later, have an outsized affect on an consequence. He anticipated code reds will probably be a norm as the corporate hopes to achieve distance from the likes of Google and DeepSeek.
“My guess is we’ll be doing these as soon as, perhaps twice a 12 months, for a very long time, and that’s a part of actually simply ensuring that we win in our area,” Altman mentioned. “A number of different firms will do nice too, and I’m blissful for them.”
This story was initially featured on Fortune.com
