‘Mark’ ⇥ colossus.com

I think Meta is a perversely interesting company which means this week I watched, as I do every year, its Connect presentation. It is once again a glimpse into what Meta aspires toward — Apple-quality hardware matched with Google-quality services — compared to what it actually does: track users’ digital and physical activities to sell ads.

Said hardware this year was interesting. In addition to a whole smattering of eyeglasses that look like slightly chunkier versions of classic glasses styles, there are now models without a camera.¹ In an interview with Joanna Stern, Mark Zuckerberg avoided answering a question about whether they were created in response to overall privacy concerns about cameras on faces by turning it into a question about whether they were developed in the past two months to address recent criticism. But questions about the ethics of wearable cameras in normal-looking glasses have been around since they were released three years ago.

Meta also showed off a much slimmer and more normal-looking virtual reality headset; Zuckerberg compared the glasses’ approximately 100-gram weight to that of Apple’s Vision Pro. They will apparently ship in “Spring” for $1,300 USD, which means you could buy three of them for only a little more than a single Vision Pro. (For what it is worth, in a recent interview with Nilay Patel, Mark Gurman said he still thinks Apple’s glasses are coming next year.)

Finally, there was a prototype gadget that looked a little bigger than the body of an Apple Watch and which Zuckerberg says is designed to live on a keychain. It supposedly allows you to use the company’s Muse A.I. assistant in something that is not its glasses or a smartphone — the latter because Meta does not have a smartphone operating system and, I guess, does not want to develop a companion app exclusively for Japan. I feel like I have seen this before.

But as I watched the presentation, I was mainly thinking of Jeremy Stern’s very long profile of Zuckerberg in Colossus. Specifically, this part, a summary of Stern’s conversations with Meta Superintelligence Labs’ “competitors and investors who agree to speak to me on background”:

AI is not, in fact, God. Instead, it does math and solves a limited set of problems humans face, and is otherwise simply useful and cool. Anthropic, and to a lesser extent OpenAI, have trouble ever accepting this fact. Zuckerberg does not. He has not spent a decade comparing his company to the Manhattan Project, and thus he is not above pushing the frontier of AI to help people book airline tickets, make dinner reservations, and edit photos. He will use it to drive down the cost of serving his users to zero, and to drive up his revenue by improving ads. He will use cash from the ad business — and his ownership of data centers, chips, and other infrastructure that Anthropic and OpenAI have to pay to rent — to undercut them on price. The potential install base for his AI is 3.6 billion people, who don’t care whether a given model is six months behind the frontier.

If AI commoditizes, then Anthropic and OpenAI go to zero, and value accrues instead at the complements Meta already dominates, like distribution, attention, personalization, and commerce. If it doesn’t commoditize, then at least he is not his competitors’ prisoner the way he’s been with Apple, and all he has to do is remain within six months of the frontier, which he’s already close to. Heads, he wins; tails, he wins.

I cannot see myself buying anything Meta releases; as curious as I have become with the company, I find it fundamentally repulsive. This part of the article from Stern, though, is a perfect encapsulation of why, regardless of having little success outside its core advertising business, Meta will keep going. My interpretation of this is that Meta has somehow harnessed controversy as a positive. Not only will people continue using its services regardless of how many bad news stories Meta endures, those very problems mean it is underestimated, so it can keep trying stuff and seeing what works.