Meta Has Lost More Than $85 Billion on Its Glasses and Headsets. Here's What Would Stop It.

Meta Platforms (NASDAQ:META) spent Wednesday and Thursday showing off hardware. At its two-day Connect event, the company introduced a refreshed lineup of artificial intelligence (AI) glasses, including a $449 third-generation Ray-Ban model, a $349 pair with no camera at all, and $1,300 virtual reality glasses due next spring.

The keynote didn't dwell on what all of that hardware costs shareholders. Reality labs, the segment housing Meta's glasses and headsets, lost $4.6 billion in the second quarter on just $431 million of revenue. And since the end of 2020, the segment's operating losses add up to more than $85 billion.

Investors don't seem too bothered. Shares are trading near $775 as of this writing, after touching a fresh 52-week high Thursday. Still, what would it take to stop the losses?

Image source: The Motley Fool.

Slowing losses

Reality labs' annual operating losses have climbed every year this decade -- from $10.2 billion in 2021 to $13.7 billion the next year, then $16.1 billion, $17.7 billion, and finally $19.2 billion in 2025. But the growth of those losses has been slowing the whole way. The annual loss grew 35% in 2022, 18% in 2023, 10% in 2024, and just 8% last year. And through this year's first half, the segment lost $8.6 billion, slightly less than the $8.7 billion it lost over the same stretch of 2025.

In other words, the annual loss was still getting bigger through 2025, but barely.

The second quarter showed why. The segment's revenue rose 16% year over year, driven by strong AI glasses sales, partially offset by lower Quest headset sales. Meanwhile, its costs and expenses grew just 3% year over year -- in a quarter when Meta's total costs and expenses jumped 55%.

The tech company is spending aggressively almost everywhere else. Its hardware unit has become the exception.

Management says this is the peak

That restraint is deliberate. On the company's fourth-quarter earnings call in January, CEO Mark Zuckerberg said Meta is directing most of its reality labs investment toward glasses and wearables going forward.

"I expect Reality Labs losses this year to be similar to last year, and this will likely be the peak as we start to gradually reduce our losses going forward," Zuckerberg said.

Chief financial officer Susan Li said the same on the call, guiding for reality labs operating losses "remaining similar to 2025 levels."

Guidance like that can change, of course. But two quarters in, the plan is holding. First-half losses came in a touch below last year's, and if the second half tracks the first, 2026 could be the first year since at least 2021 in which the segment's loss doesn't grow.

Could the losses disappear?

Stopping the growth of the losses is one thing. Erasing them is another.

The gap is enormous: reality labs generated $2.2 billion of revenue in all of 2025 against $21.4 billion of costs and expenses. Revenue would need to grow nearly tenfold, with costs held flat, just for the segment to break even.

The glasses business, still in its early innings, is the one line moving in the right direction. Zuckerberg said in April that the number of people using Meta's AI glasses daily had tripled over the past year, calling the category one of the fastest-growing in consumer electronics history. And this week's lineup, with prices now running from $249 to $1,300, may widen the pool of buyers.

However, at $431 million of quarterly revenue, I think it could take years of growth like this before the losses shrink meaningfully.

In the meantime, the cost lands on shareholders. Over the past 12 months, reality labs has lost about $19 billion from operations.

That comes to about $7.50 per diluted share before taxes -- call it $6 per share after. With the stock priced at about 22 times the earnings analysts expect for next year, that lost profit is arguably worth more than $100 per share.

Ultimately, what would stop these losses from growing is already happening. Making them disappear is a much longer road, requiring glasses revenue at many times today's scale.

I think that's a tolerable trade-off. Meta's apps generated about $102 billion of operating income in 2025, enough to pay for the hardware bet more than five times over. The drag wouldn't keep me away from the stock at today's valuation.

But the number to watch from here isn't the cumulative loss. It's whether 2026 comes in around $19 billion, the way management expects.

Daniel Sparks and his clients do not have positions in any of the stocks mentioned. The Motley Fool has positions in and recommends Meta Platforms. The Motley Fool has a disclosure policy.