The AI Industry Is Running Out of Time
The Algorithmic Bridge Alberto Romero ● Covered by 2 sources
Anthropic and OpenAI are both quietly prepping IPOs while burning cash at massive scale. The author argues this isn't about AGI arriving — it's about insiders cashing out before the hype fades.
Based on reporting by The Algorithmic Bridge, Alberto Romero — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Anthropic just confidentially filed a draft S-1 with the SEC, putting it on a path toward a public listing. The company's annualized revenue run rate is reportedly closing in on $50 billion, and it may post its first-ever profitable quarter soon. OpenAI is running roughly parallel, though with lower revenue and no profitability in sight. Both are still burning enormous amounts of cash, and both are apparently getting ready to ask the public market to help foot the bill.
That timing is the interesting part. Companies can lose money for years without much scrutiny — Uber did it for a decade — but these two are operating at a scale where the losses are tied to planetary-level energy use and jaw-dropping capital expenditure with no clear payoff yet. The piece argues that both firms have become too big to fail, which means their financial fragility is no longer just their problem. If they stumble, the fallout spreads. An IPO, in this reading, isn't a victory lap toward AGI. It's a mechanism for early investors and insiders to offload accumulated risk onto pension funds, index funds, and everyday retail investors before the music stops.
The cracks are already showing outside the hype bubble. Microsoft reportedly dropped its Claude Code licenses for an internal tool because the costs didn't add up. Uber capped employee AI coding spend at $1,500 a month. Starbucks pulled an AI tool from North America after nine months of reliability headaches. Even Sam Altman has publicly called AI compute costs a serious problem. None of these are one-off glitches — they point to a widening gap between what AI vendors promise and what enterprises are actually getting for their money.
The argument isn't that Anthropic and OpenAI executives are cynically engineering an exit scam. It's more that Silicon Valley's belief in imminent superintelligence is genuine, but that belief doesn't change who absorbs the downside if the timeline is wrong. Whether AI eventually delivers transformative value or turns out to be overhyped, the people running these companies come out ahead — as pioneers of a new era or as newly minted billionaires. The rest of us only come out ahead if the technology actually works, reliably, at scale, which right now remains an open question given how stubbornly these systems keep hallucinating.
My take — AI-written commentary, not fact-checked reporting
I write about AI because I think it's genuinely important, not because I think its current stewards are trustworthy stewards of that importance, and this piece nails the asymmetry: OpenAI and Anthropic win regardless of outcome, while the rest of us only win if the tech actually delivers. That's not a conspiracy, it's just how unregulated financial engineering around 'inevitable' technologies always plays out, and Europe's instinct to ask hard questions before writing blank checks looks less like caution and more like basic self-preservation.
Read more about this at: The Algorithmic Bridge