Meta lost 20 million users last quarter and wants to spend billions more on AI

Meta had a rough quarter by user numbers, and the company’s response is basically: don’t worry, here’s more AI spending.

On Wednesday’s earnings call, Meta disclosed that its “Family daily active people” — that’s the combined daily user count across Facebook, Instagram, WhatsApp, and Messenger — fell by 20 million compared to the previous quarter. That’s not a rounding error. That’s a lot of people who just stopped showing up.

Meta’s official explanation? Internet disruptions in Iran and restrictions on WhatsApp access in Russia. Maybe that’s true, maybe it’s convenient. Given how Meta has been bundling these stats together for years, it’s hard not to wonder if something bigger is shifting underneath.

Now here’s the part that makes you raise an eyebrow: despite losing users, Meta plans to pump billions more into AI investments this year. Not a little more. Billions. The logic seems to be that AI is the future, and if you throw enough money at it, users will come back or new ones will appear. I’m not so sure.

AI is certainly a powerful tool — Meta’s been using it for content recommendations, ad targeting, and even generative features inside its apps. But when your core user base is shrinking, pouring cash into AI feels a bit like rearranging deck chairs on the Titanic. Except the deck chairs are made of GPUs and the ocean is full of TikTok.

I’ve seen this pattern before. Companies hit a growth plateau, and instead of addressing the product experience or user retention honestly, they pivot to a shiny new tech narrative. AI is the perfect cover story because it’s hard to argue against “the future.” But let’s be real: if Meta can’t keep 20 million people using its apps in one quarter, no amount of AI-generated stickers is going to fix that.

What’s more interesting to me is the bundling of stats. By lumping four very different apps together, Meta can hide weakness in one platform behind strength in another. Facebook might be bleeding users, but if WhatsApp is flat, the combined number looks okay. Until it doesn’t. And this quarter, it didn’t.

I’d love to see a breakdown by app, but Meta doesn’t provide that. They never have. So we’re left guessing which platform is actually in trouble. My bet is Facebook, which has been losing younger users for years. Instagram still has cultural momentum, but even that’s under pressure from TikTok and BeReal clones.

Meta’s AI push isn’t stupid — it’s just premature. The company should probably figure out why people are leaving before spending billions on technology that might make the experience worse. More AI-driven content feeds could accelerate the enshittification that’s already driving users away.

But hey, the stock market seems fine with it. Meta’s share price barely budged. Investors love an AI story, even when the underlying business is showing cracks. That disconnect won’t last forever.

For now, Meta is betting that AI can paper over user losses. I’m skeptical. Losing 20 million users in a quarter is a signal, not noise. Ignoring it to chase the next shiny thing is a choice. We’ll see if it pays off.