Meta (META) finally has a product people are racing to download. Its Muse AI agent passed 10 million downloads in the U.S. and Canada in just over a month. ChatGPT's app took about three times as long to get there. Then Asa Fitch at Heard on the Street did the math. It doesn't come close to paying the bills.
Muse already charges. Subscriptions run $20 or $100 a month for heavy users. Suppose 100 million people sign up next year and every one of them pays $20 a month. That's $24 billion a year. Meta's capital spending next year, the money for data centers and chips, is expected to reach nearly $200 billion. So the dream case covers about 12% of the building. Ads could add more. Meta earns about $67 a year per daily user, so 100 million more users might add another $6.7 billion. Still nowhere close.
The bills are already here. Free cash flow, the cash left over after running the business and paying for the building, was negative $2.1 billion in the third quarter. FactSet expects negative $14.1 billion in the fourth. Long-term debt stood at $80.9 billion at the end of the third quarter. It was $28.8 billion at the end of 2024. On Wednesday Meta set up a new class of nonvoting shares so it can raise equity later. Investors didn't love that. The stock fell almost 7% Thursday and about 1% more Friday.
The trust problem
There's a limit even a great product can't fix. Only 8% of people in a recent survey said they'd trust Meta with a password. Google (GOOGL) got 30% and Apple (AAPL) got 23%. An agent that can't log into your email or your shopping accounts can't do much for you. Muse has around 2.5 million daily users right now. Mark Zuckerberg says Meta is preparing for a billion. Analysts' guesses run from $10 billion of revenue at 100 million users to about $50 billion at 175 million. That spread tells you how little anyone knows.
Our read
This is an Investments/Risk story, IN04. A hit app and a good investment are different things. Downloads are free to count. Cash is what pays for data centers. The numbers we watch at Meta are free cash flow, then debt, then how much the new shares would shrink your slice of the company. That shrinking is what dilution means. Muse would have to reach several hundred million users and keep them before any of those turn. We hold META at weight in 20 books. Nothing added. Google and Apple, named in the trust survey, are held at weight in 28 and 24 books. Nothing added there either.
For your own plan, the one thing to look at is position size, meaning how much of your money rides on one stock. Meta dropped almost 7% in a day this week, and it's one of a handful of names doing most of the AI spending. Open your statement and add up what you hold in those few builders, directly and through your index funds. The one check this week is whether a 7% day in that group would change how you sleep. If it would, change the mix before the next earnings call. Umbrellas cost less before the clouds show up.
