OpenAI and Anthropic IPO Analysis
Both companies may go public in late 2026. One is targeting a $1 trillion valuation on $14 billion in projected losses. The other just raised at $380 billion on $14 billion in annualized revenue. This essay breaks down what the numbers say, what they don't, and how to think about buying either stock on day one.
The Two Biggest IPOs in AI History
OpenAI and Anthropic are racing each other to the public markets. Both companies have hired IPOInitial Public Offering — when a private company first sells shares to the public. counsel, started talks with investment banks, and are building out investor relations teams. Betting markets give Anthropic a 72% chance of listing before OpenAI.
These would be among the largest IPOs ever. For context, Saudi Aramco's 2019 IPO raised $25.6 billion at a $1.7 trillion valuation. Meta's 2012 IPO raised $16 billion at a $104 billion valuation. OpenAI and Anthropic could each be larger than Meta's listing.
If you're an individual investor, you'll probably have the chance to buy shares within the first few weeks of trading. The question is whether you should.
The Numbers Behind the Hype
OpenAI went from roughly $6 billion in 2024 revenue to a $20 billion annualized run rate by the end of 2025. That growth rate is extraordinary. Google took seven years to hit $20 billion. Facebook took six. OpenAI did it in roughly two years from ChatGPT's launch.
But revenue growth is only half the story. The other half is the burn rate.
| Year | Revenue | Cash Burn | Net Loss |
|---|---|---|---|
| 2024 | ~$6B | ~$8B | ~-$5B |
| 2025 (est.) | ~$13B | ~$12B | ~-$5B |
| 2026 (proj.) | ~$20B | ~$17B | ~-$14B |
| 2027 (proj.) | ~$30B | ~$35B | ~-$18B |
| 2028 (proj.) | ~$50B | ~$47B | ~-$15B |
The cash burn is projected to increase for the next three years even as revenue grows. HSBC estimates OpenAI faces a $207 billion funding shortfall by 2030. That means the IPO isn't optional. OpenAI needs public market capital to survive.
The Enterprise-First Model
Anthropic looks different from OpenAI in one critical way: its revenue mix. Roughly 80% of Anthropic's revenue comes from enterprise clients. OpenAI's consumer product (ChatGPT) generates a larger share of its revenue, which means higher churn risk and lower contract values.
Enterprise revenue is stickier. Companies that integrate Claude into their workflows don't switch providers on a whim. That makes Anthropic's $14 billion annualized revenue arguably more durable than OpenAI's $20 billion.
Anthropic's valuation growth from 2023 to February 2026: Started at $18 billion in 2023, increased to $62 billion in March 2025, jumped to $183 billion in September 2025, and reached $380 billion in February 2026. This represents over a 20-fold increase in just three years.
Anthropic's valuation doubled in five months. That's the kind of markup that historically signals either a generational company or a private market bubble. Possibly both.
The $30 billion raise led by Coatue and GIC included investments from Microsoft and NVIDIA, both of which are also invested in OpenAI. Strategic investors hedging their bets across competitors is a signal worth noting. It means even the smartest money in the room isn't sure which company wins. For investors thinking about how to allocate capital across multiple AI bets, the framework in The AGI Portfolio sketches three portfolio structures keyed to different AGI timelines.
The Comparison That Matters
| Metric | OpenAI | Anthropic |
|---|---|---|
| Annualized Revenue | ~$20B | ~$14B |
| Latest Valuation | ~$500B (secondary) | $380B |
| Price/Revenue | ~25-50x | ~27x |
| Revenue Mix | Consumer + Enterprise | 80% Enterprise |
| Key Product | ChatGPT | Claude API + Code |
| Profitable? | No (est. 2030) | No |
| Major Backers | Microsoft, SoftBank | Amazon, Google |
| Key Risk | $79-134B Musk lawsuit | Margin compression |
Both companies are valued at roughly 27x their current annualized revenue. For comparison, Microsoft trades at about 13x revenue. Google trades at about 7x. Salesforce trades at about 8x. These AI companies are priced at 2-4x the multiples of the most expensive large-cap tech stocks.
That multiple implies the market expects both companies to grow revenue by 40-60% annually for several years while eventually achieving profit margins comparable to Google or Microsoft. That's possible. It's also the best-case scenario already baked into the price.
What Could Go Wrong on Day One
How to Think About Buying
Consider Buying If
You have a 10+ year time horizon and believe AI will be the dominant computing platform (like the internet was). You plan to buy a small position (1-3% of portfolio) and add on any significant dip. You're comfortable with the stock potentially dropping 30-50% in the first year. You believe one of these companies will become the next Microsoft or Google, and you accept the risk that neither might.
Wait If
You need the money within 5 years. You can't stomach a 50% drawdown. You're buying because of FOMO rather than a thesis on the business model. You'd be allocating more than 5% of your portfolio to a single pre-profit company. History says the best entry point for most hot IPOs is 6-18 months after listing, once the hype cools and the first few earnings reports give real data on growth durability.
How I Built This
Analysis based on public financial disclosures, private funding round announcements, investment bank research, and public reporting from CNBC, Fortune, Financial Times, and Reuters as of February 2026.
