Anthropic Or OpenAI: Can Both Win?

SPYYTD+7.9%QQQYTD+7.7%
Analyze ANT →

Today’s big story: Could Anthropic Stock Fall 25%?

Anthropic starts trading on Nasdaq today. It priced its shares at $83 on Wednesday evening, which values the company at about $1.1 trillion. OpenAI, the AI company most people think of first, still can’t be bought on any exchange.

So investors are asking who wins the AI race. In our view, a more useful question is whether there has to be only one winner.

Image from Unsplash

Is AI A Winner-Takes-All Race?

Not if it plays out like cloud computing. The last three big technology races ended three different ways.

In search, Google‘s Services business made $139.4 billion of operating profit last year on $342.7 billion of sales. That is profit from running the business, before interest and taxes. No rival comes close.

In cloud computing, Microsoft, Amazon and Google each run a cloud business with $58 billion or more of sales a year. All three make money. In AI chips, Nvidia‘s computing and networking business sells nearly 12 times what AMD‘s data center business does.

Race Leader, sales Next, sales Outcome
Search Google Services $343B none close One winner
Cloud Microsoft $138B Amazon $129B, Google $59B Three winners
AI chips Nvidia $193B AMD $17B 12 to 1

Source: Trefis, from each company’s latest annual filing: 2025 for Google, Amazon and AMD. Microsoft’s year to June 2026, including server software. Nvidia’s year to January 2026.

Which race does AI look like? So far, the evidence points to cloud. Both are built the same way. Huge computing bills are paid up front. The buyers are businesses that like having two suppliers.

Anthropic alone has signed up for $900.9 billion of computing. In our view, no single buyer would commit that much on a winner-takes-all bet that might not pay.

The big backers are hedging too, the way cloud customers do. Amazon, Google and Microsoft all own stakes in Anthropic. Microsoft is also a major investor in OpenAI.

Even Microsoft backs both labs rather than choose between them. If all your AI money sits in a single lab’s stock, a wrong guess could cost you a lot. Instead of one company, the Trefis High Quality (HQ) Portfolio owns 30 quality companies in equal amounts, and it is up more than 105% since September 2020. Look inside HQ.

The case for one winner is habit. “Google it” became a verb, and ChatGPT has a similar head start in people’s daily lives. Both pulls are real. So far, the money points to a shared market. Even a winner that shares its market could be very large. One view of how large is in Anthropic Stock To $500 In 5 Years?

When Will We Know Who Wins?

Later than most people think, if cloud is a guide. Amazon launched its cloud business in 2006. Ten years later, in 2016, it had $12.2 billion of sales. By then, the three-way shape was clear.

Third place took longer. Google Cloud lost $5.6 billion in 2020. It made its first profit, $1.7 billion, in 2023. Last year it made $13.9 billion. That turn from a big loss to a big profit took five years.

ChatGPT launched almost four years ago. On cloud’s timeline, the shape of the AI race would become clear around 2032. Profits for second and third place would come later still.

How Will OpenAI Compare With Anthropic?

OpenAI doesn’t publish its finances, so outsiders can’t measure it. Anthropic now can be measured. Its filing sets the yardstick OpenAI will face if it lists.

Three numbers stand out. Anthropic’s sales pace reached $81 billion a year in October. That is what it would take in over a full year at its October rate. It brought in $6.6 billion of cash from running its business in the first nine months of this year. And $451.6 billion of its computing commitments, bills it has agreed to pay, fall due before the end of 2030.

In our view, the lab that turns more of its sales into cash will earn the higher stock price. So watch the cash, not which chatbot is better.

How Do You Invest Without Picking A Winner?

Start with who supplies the computing. Both labs rent much of it from the same three cloud companies: Amazon, Microsoft and Google. Those three cloud businesses made $116.5 billion of operating profit in their latest years. The three also own stakes in the labs.

So if one lab wins, the cloud companies still get paid. If both win, they get paid by both. If you own an S&P 500 fund, you already own all three.

They carry risk too. They are spending heavily on this race. If both labs slow down at the same time, the cloud companies would feel it first.

In our view, a plan that needs no prediction has three parts. Hold the cloud companies through the index fund you already own. Add a small stake in the lab you believe in. Keep it small enough that a 30% fall would barely dent your savings. And avoid an all-in bet on one name. How and when to buy that stake is covered in $100K In Anthropic Stock: Buy Now Or Wait?, which looks at 31 past listings.

To sum up, search had one winner, cloud had three, and AI so far looks like cloud. Nobody will know the final shape for years.

That is why we prefer a plan that does not depend on picking the winner. We think ours holds up well, and we would gladly learn from a better one.

Found A Better Strategy? We’d Love To Learn.

Ours is the Trefis High Quality (HQ) Portfolio: 30 quality stocks in equal weight. Since its September 2020 launch it has returned more than 105%, ahead of a benchmark that combines the S&P 500, S&P MidCap and Russell 2000. If yours has done better, tell us about it. If not, come take a look.

See The HQ Portfolio

For education, not a recommendation to buy or sell.

For questions about this article or underlying research, email help@trefis.com.