Meta’s Reality Labs is a money pit that keeps getting deeper

Meta’s Reality Labs division has been a financial black hole for years, and the latest numbers don’t offer any hope of a turnaround. The company just reported another quarter where it lost billions on AR/VR hardware and software, and the situation is only going to get worse as Meta doubles down on AI infrastructure.

Let’s be blunt: Reality Labs lost something like $4.5 billion in the last quarter alone. That’s not a typo. That’s more than most companies make in a year, and Meta is just burning it on headsets that most people still don’t want to wear. I’ve tried the Quest 3, and it’s decent hardware, but the ecosystem is still thin and the use cases are niche. Gaming? Sure. Work? Not really. Social? Creepy at best.

And now Mark Zuckerberg is throwing AI into the mix. Meta’s AI spending is ramping up fast, with billions going into data centers, GPUs, and research. The company is investing in everything from large language models to AI-powered glasses, which sounds cool until you realize this is all layered on top of that Reality Labs money pit. The combined spending is staggering.

The problem isn’t that Meta has bad ideas. The problem is that it’s betting on two huge, unproven markets at the same time. AR/VR hasn’t found its killer app yet, and AI is still in the hype phase where everyone is spending but few are profiting. Meta is basically trying to sprint a marathon while carrying a backpack full of cash that’s on fire.

I get the long-term vision. Zuckerberg wants to own the next computing platform, whether it’s AR glasses or an AI assistant. But the quarterly losses are real, and investors are getting restless. The stock took a hit after this report, and I can’t blame them. At some point, you have to show that all this money is leading somewhere other than a bigger burn rate.

What’s frustrating is that Meta’s core business—social media and advertising—is still printing money. Facebook and Instagram are cash cows. But instead of returning that to shareholders or investing in safer bets, Meta is funneling it into these moonshots. That’s fine if you have the stomach for it, but the timeline keeps getting pushed out. Reality Labs was supposed to be profitable by now, and it’s not even close.

The AI spending is a bit different. Meta has actually released some solid open-source models, like Llama, and its AI research is legit. But the infrastructure costs are insane. Training and running these models at scale requires billions in hardware, and Meta is competing with Microsoft, Google, and Amazon for the same scarce GPUs. That’s a bidding war nobody wins.

So where does this leave us? Meta is a company with two very expensive hobbies: losing money on VR and spending even more on AI. The metaverse dream is still alive, but it’s looking more like a nightmare for the balance sheet. I don’t see a path to profitability in the next couple of years, and the AI bet is a high-risk gamble that could pay off or blow up.

Honestly, I’d rather see Meta focus on one thing and do it well. Pick a lane: either commit to AR/VR and accept the losses until it matures, or go all-in on AI and stop pretending the metaverse is the future. Trying to do both is just going to burn through cash faster than either bet can justify. But hey, it’s not my money.