Mark Zuckerberg: The Metaverse, Codec Avatars & Quest 3
Quick Take
A genuinely novel interview format — the first podcast conducted entirely in VR using photorealistic avatars. The technology is impressive, but Zuckerberg's claims about timeline and adoption deserve scrutiny. Two years later, we can assess which predictions panned out and which remain aspirational.
Key Claims Examined
🎭 Codec Avatars Are "Photorealistic"
"We've scanned ourselves and a lot of different expressions, and we've built a computer model of each of our faces and bodies... it can basically send an encoded version of what you're supposed to look like over the wire."
Our Analysis
The Codec Avatars demonstrated in this interview are genuinely impressive — arguably the most realistic real-time VR avatars ever publicly shown. But there are important caveats:
- What's real: The technology works. Facial expressions transfer convincingly, and viewers genuinely struggle to tell it's not video. This is a significant technical achievement.
- The catch: Creating a Codec Avatar requires hours in a specialized scanning rig with hundreds of cameras. Zuckerberg and Lex had access to Meta's research labs. This isn't something consumers can do.
- 2+ years later: As of early 2026, Codec Avatars remain a research project. Consumer Quest headsets still use the simpler, cartoonish Meta Avatars that Zuckerberg explicitly contrasts against.
- The honest framing: To his credit, Zuckerberg doesn't claim this is shipping to consumers tomorrow. He presents it as the future they're building toward.
Verdict: Technically accurate, but far from consumer-ready
🥽 Quest 3 Mixed Reality Claims
"Mixed reality is going to be really big... Quest 3 has full color passthrough at a much higher resolution... You can see the real world and virtual objects together seamlessly."
Our Analysis
Quest 3 launched in October 2023, shortly after this interview. How did the claims hold up?
- What shipped: Quest 3's mixed reality is genuinely impressive. The passthrough quality is dramatically better than Quest 2/Pro, and reviewers confirmed the step-up in quality.
- The reality check: "Seamlessly" is marketing speak. Passthrough still has visible grain, edge distortion, and latency. It's functional but not invisible.
- Adoption: Quest 3 sold well (estimated 5-10M units by end of 2024), but mixed reality hasn't become the killer app. Most users still primarily use it for VR gaming.
- Apple Vision Pro entered: Apple's 2024 headset raised the bar for passthrough quality, making Quest 3's look noticeably worse in comparison — though at 7x the price.
Verdict: Largely delivered on hardware claims
🌐 The Metaverse Timeline
"We're building toward a future where you can be present with anyone... I think this is the future of how human beings connect to each other in a deeply meaningful way on the internet."
Our Analysis
The "Metaverse" framing that Meta bet its company name on has aged awkwardly:
- The pivot: By 2024, Meta quietly de-emphasized "Metaverse" messaging in favor of AI. The company now leads with its LLaMA models and AI assistants, not Horizon Worlds.
- Reality Labs losses: Meta's VR division has lost over $50 billion since 2020. While Zuckerberg remains committed, investor pressure forced significant layoffs and strategy adjustments.
- What's true: VR/AR technology continues advancing. Quest headsets sell. But the grand vision of billions of people "living" in the Metaverse hasn't materialized.
- The Zuckerberg pattern: He's made similar "this is the future" claims before (Facebook Home, Facebook Watch, Libra/Diem crypto). His batting average on platform predictions is mixed.
Verdict: Vision remains speculative; timeline unclear
🤖 AI-Powered Social Connection
"Instead of actually transmitting a video, what it does is we've scanned ourselves... and collapsed that into a codec that then when you have the headset on your head, it sees your face, it sees your expression, and it can basically send an encoded version."
Our Analysis
The AI/ML technology underlying Codec Avatars is genuinely sophisticated:
- The innovation: Using machine learning to compress facial expressions into a "codec" that can be transmitted efficiently is clever engineering. It solves real bandwidth problems for realistic VR communication.
- What's not said: This approach requires extensive pre-scanning to build a personalized model. Scaling this to billions of users is a massive infrastructure challenge that hasn't been solved.
- The AI shift: Ironically, Meta's biggest AI success since this interview hasn't been in avatars — it's been LLaMA and generative AI tools. The AI investments that matter ended up being elsewhere.
- Bandwidth efficiency claim: This is technically accurate. Sending expression parameters is far cheaper than streaming 3D video. The question is whether the scanning cost makes it worthwhile.
Verdict: Technically sound approach, scaling challenges unaddressed
🏠 The "Presence" Revolution
"It really feels like we're in the same room... This is the most incredible thing I've ever seen."
Our Analysis
Lex's enthusiasm is genuine but worth examining:
- The setup: This interview was orchestrated as a tech demo. Meta controlled the environment, lighting, and presentation. It was designed to impress.
- First impressions vs. sustained use: Many VR experiences are mind-blowing initially but lose novelty. The question isn't "is it impressive?" but "will people use this regularly?"
- VR adoption reality: Despite impressive demos, VR headset usage drops significantly after the first month of ownership. The "presence" doesn't outweigh the friction of wearing a headset.
- The interviewer's role: Lex is clearly impressed, but he's also getting exclusive access to unreleased tech with Meta's CEO. The incentives favor enthusiasm over critical examination.
Verdict: Impressive demo ≠ mainstream adoption
What Should We Believe?
Mark Zuckerberg remains one of the most consequential technology executives, and this interview showcases both Meta's genuine innovations and its pattern of over-promising on timelines:
- The technology is real: Codec Avatars, high-quality passthrough, and spatial computing represent genuine advances. Meta's Reality Labs has produced impressive research.
- The timeline is always "a few years away": Zuckerberg has been promising the Metaverse is imminent since 2021. Each year, the vision remains compelling but consumer adoption stays niche.
- The pivot tells the story: Meta's 2024 embrace of AI over Metaverse messaging suggests even they recognize the original timeline was too aggressive.
- Demo ≠ Product: This interview showcases research-lab technology that requires hours of scanning and millions in infrastructure. The gap between demo and product is years, possibly decades.
- Lex as interviewer: The interview is engaging but not adversarial. Questions about Reality Labs' losses, layoffs, or the Metaverse skepticism in the market go unasked.
The Bottom Line
This is the most technically impressive podcast episode ever recorded — conducting an interview in VR with photorealistic avatars is a genuine first. The technology Zuckerberg demonstrates is remarkable and represents real engineering achievement.
But impressive demos have never been Meta's problem. Execution, adoption, and timing have been. The Codec Avatars shown here remain research projects in 2026. The Metaverse vision that once defined the company has been overshadowed by AI. Quest headsets sell to enthusiasts, not the mainstream.
Watch for the technology showcase. Be skeptical of the timeline claims. And remember: if Zuckerberg's predictions were reliable, we'd all be living in the Metaverse by now.