The AI giant is juggling a messy IPO push, executive drama, and a CFO who’s basically waving a red flag. Here’s what’s really going on.
The Vibe Check Nobody Asked For
Let’s be honest. OpenAI is having a moment, and not the good kind.
The company that practically invented the modern AI gold rush is now making headlines for all the wrong reasons. Executive reshuffles. Cancelled projects. A CFO who reportedly thinks the company isn’t ready to go public. And a CEO who seems determined to go public anyway.
It’s a lot. And it’s happening fast.
Just a few weeks ago, OpenAI closed a jaw-dropping $122 billion funding round at a post-money valuation of $852 billion. That’s almost a trillion dollars. On paper, that sounds incredible. But dig a little deeper, and the picture gets complicated, fast.
The IPO Drama Nobody Can Stop Talking About
Here’s the thing about OpenAI’s IPO plans: they’re moving forward even though the company’s own CFO has serious doubts.
Sarah Friar, OpenAI’s Chief Financial Officer, has reportedly told associates that the company won’t be ready to go public in 2026. She’s flagged concerns about compliance, internal systems, and the company’s eye-watering spending commitments. According to reports, OpenAI’s cash burn could exceed $200 billion before it ever reaches positive cash flow.
That’s not a small concern. That’s a five-alarm fire.
And yet, CEO Sam Altman is reportedly pushing hard to get the IPO done as soon as possible. The tension between the two is real. According to Pivot to AI, Altman has even excluded Friar from key financial conversations — including a major meeting about server spending with one of OpenAI’s top investors. Her absence was reportedly “noticeable and awkward.”
That’s not a great sign for a company about to ask the public to trust it with their money.
Wait, Who Does the CFO Even Report To?
Here’s where it gets genuinely weird.
Normally, a CFO reports directly to the CEO. That’s just… how companies work. But at OpenAI, Friar reports to Fidji Simo, the CEO of AGI deployment, not to Altman. And Simo? She’s currently on medical leave.
So the CFO reports to someone who isn’t there. And the CEO is leaving the CFO out of financial meetings. This is the company that wants to go public.
The Verge put it plainly: the vibes are off. And honestly? That’s an understatement.
The $122 Billion That Isn’t Quite $122 Billion
Let’s talk about that funding round, because it’s not as clean as it sounds.
Yes, OpenAI raised $122 billion. But here’s the catch, only about $25 billion of that is immediately usable cash. The rest is conditional, staged, or tied to specific milestones like an IPO or achieving AGI.
And the investors? They’re not exactly neutral parties. Amazon put in $50 billion, but that’s tied to an eight-year, $100 billion AWS usage contract. Nvidia invested $30 billion, and analysts widely expect those funds to cycle right back into Nvidia chip purchases. SoftBank’s $30 billion comes in tranches, tied to OpenAI’s technical progress and IPO timeline.
The ELEC calls this “circular financing.” Companies invest in each other while simultaneously buying each other’s services. It’s a structure that can make revenue look bigger than it really is. Analysts have compared it to the dot-com bubble of the early 2000s, when companies inflated revenues through mutual advertising deals.
To justify its $852 billion valuation, OpenAI would need to generate $300 billion in annual revenue with a 30% operating margin within a decade. For context, Google and Apple took years to hit those levels, and they weren’t burning $178 million per day.
The Executive Exodus Nobody Saw Coming
The financial drama is only half the story. OpenAI’s leadership has been reshuffling at a dizzying pace.
Just last week, the company announced a string of C-suite changes. COO Brad Lightcap stepped away from his role to focus on “special projects.” CMO Kate Rouch departed to focus on her health. Fidji Simo, who had been running the applications business, stepped back due to medical leave. Company president Greg Brockman stepped in to run the product organization.
That’s a lot of movement for a company that’s supposedly firing on all cylinders.
And the product side hasn’t been smooth either. OpenAI unexpectedly discontinued Sora, its AI video-generation app. It exited its Disney partnership so abruptly that Disney reportedly found out just 30 minutes after the decision was made. The company also shelved plans for a ChatGPT “sexting” feature.
“We cannot miss this moment because we are distracted by side quests,” Simo reportedly told employees before her leave. The company is now pivoting hard toward enterprise and coding tools.
The Competition Is Closing In
Here’s the part that should really worry OpenAI investors: the competition isn’t standing still.
Anthropic is eating OpenAI’s lunch in enterprise markets. In secondary markets for private stocks, demand for OpenAI shares is reportedly weak, institutional investors have tried to sell about $600 million worth of OpenAI shares and couldn’t find a single buyer. Meanwhile, Anthropic shares are surging, with about $2 billion in capital waiting to be deployed.
Goldman Sachs has reportedly waived fees when offering OpenAI shares to wealth management clients. For Anthropic? They’re charging the standard 15–20% carry. That contrast says everything.
Google’s Gemini is deeply integrated into Google’s ecosystem. Meta is investing heavily in open-source AI. DeepSeek came out of nowhere and rattled the entire industry. OpenAI’s once-unassailable lead is narrowing, and fast.
The Pentagon Deal That Backfired
Let’s not forget how this rough patch started.
Back in February, OpenAI agreed to a sweeping Pentagon contract, one that competitor Anthropic had refused to sign over concerns about autonomous weapons and domestic mass surveillance. The move sparked controversy both inside and outside the company.
Even Altman admitted OpenAI had come off as “opportunistic and sloppy.” That’s a rare moment of self-awareness from a CEO who usually projects supreme confidence.
Then came The New Yorker piece that expanded on years of reports about Altman potentially misleading OpenAI’s board and former executives. And later this month, OpenAI is set to face Elon Musk in court — a lawsuit that has already surfaced extensive internal communications from the company’s early days.
Can OpenAI Pull It Together?
Here’s the honest answer: maybe.
OpenAI still has nearly 1 billion users. ChatGPT is still the most recognized AI brand in the world. The company expects about $24 billion in revenue this year. And it has some of the best AI researchers on the planet.
But the pressure is immense. Infrastructure costs are two to three times revenue. Net losses are expected to exceed $40 billion. Profitability isn’t projected until after 2030, and that’s the optimistic scenario.
OpenAI even acquired TBPN, a viral online news show, in what looks like an attempt to control its own narrative. Simo wrote that the company needs to move beyond the “standard communications playbook.” That’s a polite way of saying: things are messy, and we know it.
“We have a strong leadership team focused on our biggest priorities,” an OpenAI spokesperson told The Verge. “We’re well-positioned to keep executing with continuity and momentum.”
That’s the official line. Whether Wall Street buys it, literally, remains to be seen.
The Bottom Line
OpenAI is at a crossroads. It’s the most talked-about AI company in the world, sitting on a mountain of funding, and racing toward an IPO that its own CFO thinks is premature. The executive drama is real. The financial structure is complicated. The competition is fierce.
This isn’t a company in freefall. But it’s also not the smooth, unstoppable juggernaut it once appeared to be. The cracks are showing. And with an IPO on the horizon, every crack is going to be examined under a very bright light.
Buckle up. This story is far from over.
