Management has stopped promising to spend ever more on its virtual reality dream, a quiet pivot that shifts the entire investment case to a massive, and costly, bet on AI.
For years, the story inside Meta Platforms (META) was that the future would be built in the Metaverse. It was the grand, company-redefining vision. But if you listened closely to the last earnings call, the most telling signal wasn’t a new promise, but an old one that went missing. The company has quietly stopped talking about the one thing that worried investors most about its virtual reality bet.
The Billion-Dollar Promise They Stopped Repeating
Just a few quarters ago, management’s guidance on the Metaverse was explicit and painful. On the costs for its Reality Labs segment, the company was clear, stating its expectation that “2024 operating losses will increase meaningfully year-over-year.” It was an explicit acknowledgment that near-term investments in the unproven digital ecosystem would weigh on margins. That language has now softened. That language has now vanished. In its place is a far more sober objective: “we are focused on making our VR business sustainable as we invest more in other areas.”
That shift from increasing losses to “sustainability” is a quiet earthquake. The Reality Labs segment, after all, is a rounding error on the top line, generating just 1% of the company’s revenue and growing at a sluggish 3% over the past year. The old promise was to pour ever more money into a tiny, slow-growing business. This strategic shift reflects an emphasis on capital discipline regarding the Metaverse. By reallocating marginal investments to reinforce its primary AI engine, management is executing a pragmatic pivot visible in its updated CapEx projections.
Where the Money Is Actually Going Now
So where has the company’s center of gravity moved? Straight into AI infrastructure, and the scale of the pivot is staggering. While the talk around Reality Labs has gone quiet, the spending plans for AI are screaming. Management is now guiding for 2026 capital expenditures to be in the range of $125 billion to $145 billion, a massive increase to build out the compute power for its new grand vision: delivering “personal super intelligence to billions of people.”
The contrast tells the whole story. The focus has shifted from a 3% growing hardware project to the AI systems that are already powering the core business, which saw overall revenue grow 26.2% last year. The company is no longer asking you to fund a growing loss center; it’s asking you to fund a massive expansion of the very engine that’s driving its current success.
A Problem Repackaged, Not Buried
This silence is reassuring. It signals a welcome, if unannounced, dose of capital discipline on the Metaverse project that has long been a drag on the stock’s valuation. By quietly capping the moonshot to fund the main engine, management is making a more pragmatic bet. The company is redirecting its focus from a speculative, low-growth hardware business to the AI investments that are already delivering measurable results in its core advertising machine. While Meta still posted a $19.1 billion operating loss in Reality Labs last year, the strategic narrative and future CapEx growth have entirely shifted to AI.
The one number to watch over the next year or so is the Reality Labs operating loss. If that figure stabilizes or shrinks, it will confirm that this pivot to sustainability is real. If it continues to balloon, the new narrative will be just talk. The company’s management has explicitly stated that 2026 losses for Reality Labs will likely match 2025 levels before any gradual reduction occurs.
What You Own Now Is Not What You Bought
This is easy to miss, but the investment you own has changed shape. Meta Platforms is no longer primarily the Metaverse company it renamed itself to be. It has quietly become an AI infrastructure company, funded by a social media empire, and you had to listen for the promise they stopped making to see it clearly.
You Cannot Catch This on Your Own
The hard part was never reading this one story; it is realizing the same quiet migration is underway beneath every name you hold, and most of it stays invisible unless you go looking. The figures that ground it for Meta Platforms are the segment breakdown. No one can audit all of that every quarter. That is precisely what the Trefis High Quality Portfolio systematizes, weighing forward-looking fundamentals across 30 names with sizing discipline and a record of outrunning the S&P 500, S&P Mid-cap, and Russell 2000.
