Latest / Tech IPO Conversations with Fexingo: Public Software Companies and Capital Markets / Why OpenAI Stays Private In A Public Market Rush
Transcript
- Lucas: If these conversations are useful for what you're building or running, I want to mention that we keep this network ad-free because of listener support. Luna: Right, so if you value deep dives into business mechanics, a small monthly contribution helps us keep producing this volume. Lucas: You can find that at buy me a coffee dot com slash fexingo. It’s not about perks, just keeping the lights on for three hundred daily shows. Luna: Thanks for that. Now back to the news—Sam Altman just told TechCrunch that it would be ill-advised for OpenAI to go public in 2026. Lucas: That is a massive signal, especially given the current market environment here on September twelfth, twenty twenty-six. We have seen NVIDIA down five percent this week and Shopify sliding eleven percent, so public markets are clearly punishing high-growth tech narratives right now. Luna: Exactly. So Altman is essentially saying the public markets aren't ready for the specific capital structure AI requires? Lucas: I think it goes deeper than just readiness. The core issue is the tension between massive, unpredictable capital expenditure and the public market's demand for predictable, quarter-over-quarter growth. When you are building frontier models, your R&D spend can swing wildly based on breakthroughs or dead ends. Luna: And the public markets hate uncertainty in their cost structures. They want to see efficiency, not just brute-force compute spending. Lucas: Precisely. Look at the broader context today. Meta is up five percent, but that is because they have optimized their ad engine and shown margin expansion. OpenAI’s model is fundamentally different—it is a research engine first, a product second. You cannot really split those two things out cleanly for a P/E ratio calculation. Luna: So staying private allows them to ignore the quarterly earnings cycle entirely? Lucas: It allows them to answer to investors who understand long-horizon horizons. But here is where it gets interesting. If OpenAI stays private, who actually owns the upside? That capital comes from giants like Microsoft and SoftBank, who are looking for strategic control, not just financial return. Luna: Which creates a weird dynamic where the company is technically independent but financially tethered to a few mega-caps. Lucas: Right. And that brings us to Anthropic. Their CEO just outlined a plan to slow down AI development. Why would they do that if they were chasing an IPO path? Luna: Because they are also privately funded and likely prioritizing safety and sustainable scaling over aggressive revenue grabs that might spook public investors. Lucas: But there is a counter-narrative happening in the robotics space. Mecka AI just nears a five hundred million dollar valuation in a Sequoia-led deal specifically for robot training data. Luna: That is huge. Robot training data is the next bottleneck after large language models. Why is that attracting such heavy private capital right now? Lucas: Because it is tangible. You can measure the physical world. Large language models are becoming commodities, but the ability to train robots to interact with physical environments is still a scarce asset. And private equity loves scarcity because they can lock up that value for ten years without public scrutiny. Luna: So while the AI software giants stay private to manage expectations, the AI hardware and data layers are getting pumped up by venture capital. Lucas: I would argue the reverse is happening. The software giants are staying private to avoid the scrutiny of public markets during this volatile period. We saw ARKK down three percent today, showing that innovation-focused ETFs are struggling to find stable ground. Luna: But doesn't staying private eventually create a liquidity problem for early employees and founders? Lucas: It does. That is why we see secondary share sales more often now. Companies like Coinbase and Roblox trade publicly, but many of their pre ipo holders use secondary transactions to cash out. It is a way to provide liquidity without going public. Luna: So the private market is acting as a holding pattern until the tech sector stabilizes? Lucas: Or until the narrative shifts. Here is the thing: if interest rates drop, or if inflation data cools further, the pressure to go public increases. Right now, with the ten-year Treasury yields fluctuating, private valuations are being scrutinized heavily. Luna: Valuation cuts are becoming common. So founders are choosing to stay private rather than take a haircut in a weak market. Lucas: Exactly. And that protects them from activist investors. In the public market, if you miss one earnings call, your stock can get hammered. In the private market, you just talk to your board. For a company like OpenAI, whose mission involves complex ethical considerations, that privacy is a shield. Luna: It also means less transparency for the average investor. We don't know exactly how much OpenAI is spending on compute versus talent. Lucas: Correct. And that opacity is risky. If you are an investor in a private AI firm, you are betting on the team's judgment, not just the numbers. It is a higher-trust, higher-risk environment compared to buying shares in Apple or Microsoft. Luna: But those public tech giants are also investing heavily in AI. So are they effectively outsourcing the risk to private startups? Lucas: Partially. They are buying the output. Microsoft invests in OpenAI, Amazon invests in various cloud AI services, but they keep their own core businesses public. They get the AI upside without exposing their entire enterprise to ai specific volatility. Luna: It is a hedge. Use the private sector for experimentation, use the public sector for monetization. Lucas: That is a very clean way to put it. And look at Tesla today. They are up three percent, and Elon Musk is unveiling the new Roadster soon. But Tesla is already public. They have to balance innovation with manufacturing margins every single quarter. Luna: Whereas OpenAI can focus purely on the model architecture without worrying about gigafactory yield rates. Lucas: Exactly. The business models are too divergent to force a public listing right now. Unless, of course, OpenAI pivots to become a pure utility provider, like an ai as a service company with steady, recurring revenue. Luna: But then they lose the 'frontier' premium. Investors pay extra for the cutting edge, which is inherently unstable. Lucas: True. So the question becomes: when does the cutting edge become the standard? Once it does, the IPO window opens. Until then, Sam Altman’s advice holds weight. Luna: Ill-advised to go public now, but perhaps inevitable later. It’s a timing game. Lucas: And a testament to how much power private capital still holds in shaping technological direction. Luna: We will have to watch the next funding round announcements closely to see if that changes. Lucas: Agreed. One thing to consider is whether the regulatory environment for AI will force more transparency, regardless of public status. Luna: Government oversight could act as a proxy for shareholder pressure, demanding audits and safety reports. Lucas: That is a fascinating angle. If the SEC or FTC starts treating AI developers like public utilities, the distinction between private and public blurs. Luna: Then maybe Altman is wrong about the timeline. Regulation might force the hand sooner than the market does. Lucas: Possibly. But for now, the silence from OpenAI regarding an IPO speaks louder than any prospectus ever could. Luna: Silence is definitely a strategy. Let’s see if the market rewards that patience or punishes the lack of liquidity. Lucas: We will know once the next earnings season hits for the public AI proxies. Until then, the private giants hold the cards. Luna: And those cards are worth billions. Lucas: Indeed. It raises the question of whether true innovation can survive under the microscope of public quarterly reporting. Luna: Or if it needs the shelter of private equity to reach maturity. Lucas: A compelling dilemma for any builder watching this space. Luna: Let’s see which side wins out in the end. Lucas: Stay curious.