The metaverse never happened, but Mark Zuckerberg’s got some funky glasses

This story originally appeared in The Technology Letter and is republished here with permission.

Oh, the Metaverse!

Remember that bit of nonsense, from three years ago? Deep in the pandemic lockdowns, it seemed we all would be living in a digital cave for the rest of eternity. The Metaverse was the signature of Facebook’s transformation into Meta Platforms.

What actually came about were waves of layoffs and massive share buybacks, which this year have boosted the stock 60%.

The Metaverse stuff is now very dated, and has been replaced with some other ambitions.

On Thursday, the company, at its developer conference showed off a cheaper version of its virtual reality headset, the “Quest 3S,” which is $300 versus the original Quest 3, introduced earlier this year, at $500.

And, the show-steeler, a prototype of a pair of glasses, code-named “Orion,” that project holograms onto one’s field of view, so that you can see your email floating in front of your surroundings. A picture of Nvidia CEO Jensen Huang demonstrating the glasses (up top) struck me as hilarious in a sublime way.

“On the negative side, the new glasses appear to be several years (and $10s of billions) away from general release (size, tech, and cost improvements),” writes Merrill Lynch’s Justin Post in his recap of the event. “Also, the event did not include any Metaverse related user metrics, suggesting limited traction for the VR ecosystem.” I’m not surprised.

An expensive demo, an unfulfilled Metaverse—Meta has become the Queen of Vaporware. But it’s okay, because Post is bullish on the shares nonetheless. He raised his price target to $630 from $563. That is a mere 11% gain from Thursday’s $567.84, but he rates the stock a Buy.

Meta stock trades for twenty-three times next year’s earnings per share currently, and Post thinks it can inch up to twenty-five times based on its artificial intelligence potential. “The company appears to be successfully innovating around new AI capabilities, driving usage growth which can offset terminal value concerns, and we see Meta as the top AI pick in consumer Internet.”

It is true that Meta has lead the move to AI technology that is “open-source.” Its Llama version of “generative” AI is all the rage, even as OpenAI continues to lose top talent.

So, Meta is very relevant technologically, just not in the fluffy way that Mark Zuckerberg had presented three years ago. The thing that’s been best for this company is driving the advertising engine, laying off staff, boosting operating profit margins, and buying back a ton of stock—the plain old stuff that doesn’t require a Metaverse.

A MORE INTERNATIONAL REDDIT

Meanwhile, the newer darling of social media, Reddit, gets a vote of confidence on Thursday from Benjamin Black of Deutsche Bank, who notes that the company is deploying localized versions of the property in Spanish, Portuguese, German, Italian, Filipino and dozens of other languages. When this happened with French-language support earlier this year, notes Black, it lead to a surge of activity among users in France of Reddit.

“This follows the initial launch of French translation earlier this year, which propelled France to be one of the fastest growing countries for Reddit, outpacing the US, which grew users 59% y/y and revenue 39% y/y in 2Q24,” he notes. “In our view, this could fast-track both content consumption, as well as content creation across the globe to the benefit of all users,” writes Black, “which should fuel the flywheel of more content attracting more users, which in turn should generate more content, attracting more users.” That is important because Reddit has only half its users coming from outside the U.S., whereas Meta, for example, gets 91% of people from outside the country. If Reddit can close that gap, muses Black, its number of users—as measured by “daily average users,” or, DAUs, the thing the Street cares about—could be 50 to 150% higher than current estimates come 2026.

I’m fairly confident that by 2026, no one will remember what they thought the DAU numbers were supposed to be for Reddit. Nor will they even remember it by the end of this year. Stuff such as DAUs, user counts, engagement, etc., all get sweated over quarter by quarter. So, while Black could be right, I doubt it will mean much for the shares on the way to 2026. Reddit has doubled since its March 21st IPO.

ROBO-TAXIS ARE THE FUTURE?

You may recall that Tesla is expected on October 10th to unveil its “robo-taxi,” an event eagerly anticipated since the company’s disappointing second-quarter report in July. The shares are up almost 30% since that report as investors have become a little more comfortable with Tesla’s pace of sales, and as they anticipate the taxi unveiling.

But just how viable is the robo-taxi thing? A big believer is Bernstein’s Nikhil Devnani, who on Thursday relates his personal trip in a Waymo self-driving taxi in San Francisco. This is by now a somewhat familiar story—Waymo, backed by Alphabet’s Google, is serving 50,000 paid rides per week in California, according to Devnani. But the “unit economics” are still rough even when not paying a driver. The Waymo car, a Jaguar I-Pace, costing $75,000, must be outfitted with as much as a $100,000 worth of LiDAR and other sensors, notes Devnani.

In a previous report—this is the third in a series Devnani has written—he had noted that “there are real challenges to scaling,” getting enough of these expensive cars on the road to have a fleet that’s economical. In the latest note, after enjoying some Waymo rides in San Francisco, Devnani concludes, “It’s early days, but recent developments and continued investment across the space leave me increasingly convinced that AVs are the future, not just for robotaxi supply but also with respect to personal car ownership in the long-run, which I think is the bigger picture end-game here.”

Perhaps, but one thing needs to be reiterated again and again. Waymo, and all other operations, such as General Motors’s Cruise business, have very, very limited deployments in areas such as San Francisco that have been extensively mapped by multiple companies. As critics of self-driving tech, such as Meta’s head of AI, Yann LeCun, have repeatedly noted, this technology does not yet understand how to drive on any road in any city in any country. For that reason, I suspect that no matter what comes from Elon Musk on the 10th of October, we are farther from the promise of self-driving than Waymo and others will admit.

This story originally appeared in The Technology Letter and is republished here with permission.