Naveen Rao: 4D Computing, AI's Energy Wall & Beating Biology
Episode
22 min
Read time
2 min
Topics
Productivity, Remote Work, Startups
AI-Generated Summary
Key Takeaways
- ✓AI Energy Crisis Timeline: Global AI energy demand is accelerating toward a hard ceiling within roughly three years. Google alone processes 3.2 quadrillion tokens monthly, consuming an estimated 12 gigawatts — nearly 30% of total worldwide data center energy today. As models scale and demand grows, current infrastructure cannot keep pace without radical efficiency breakthroughs.
- ✓Biology as Engineering Benchmark: A squirrel's brain runs on 8 milliwatts — meaning a smartphone could power over 100 squirrel brains simultaneously. GPUs move nearly 30 trillion bits per second in and out of memory; the human cortex moves only 16 billion. Reducing data movement, not increasing raw speed, is the primary lever for efficiency gains.
- ✓4D Computing Architecture: Unconventional AI's chip eliminates the traditional von Neumann memory-compute separation entirely. Each computing element stores its own state, removing the memory interface bottleneck. The architecture uses three physical dimensions via die stacking plus time as a fourth dimension, producing a chip that generates images at roughly 500 nanojoules versus millijoules on a GPU.
- ✓Sparsity as a Performance Multiplier: Reducing connections between computing elements — called sparsity — does not degrade system performance; it improves it. Sparse dynamical systems become more trainable and more scalable simultaneously, avoiding the n-squared connection scaling problem. This allows the architecture to grow without proportional energy or complexity costs, a combination previously considered unachievable.
- ✓Jevons Paradox and Market Scale: Dropping the cost of AI compute by 1,000x will not reduce total consumption — it will trigger demand expansion far exceeding the cost reduction, per Jevons Paradox. Rao projects this creates the largest market in human history, shifting infrastructure from centralized gigawatt data centers toward distributed, smaller facilities enabling billions of robotic deployments.
What It Covers
Naveen Rao, CEO of Unconventional AI, presents a new computing architecture called 4D computing that uses dynamical systems — modeled on biological brains — to achieve up to 1,000x greater power efficiency than GPUs, targeting AI's looming energy crisis within three years.
Key Questions Answered
- •AI Energy Crisis Timeline: Global AI energy demand is accelerating toward a hard ceiling within roughly three years. Google alone processes 3.2 quadrillion tokens monthly, consuming an estimated 12 gigawatts — nearly 30% of total worldwide data center energy today. As models scale and demand grows, current infrastructure cannot keep pace without radical efficiency breakthroughs.
- •Biology as Engineering Benchmark: A squirrel's brain runs on 8 milliwatts — meaning a smartphone could power over 100 squirrel brains simultaneously. GPUs move nearly 30 trillion bits per second in and out of memory; the human cortex moves only 16 billion. Reducing data movement, not increasing raw speed, is the primary lever for efficiency gains.
- •4D Computing Architecture: Unconventional AI's chip eliminates the traditional von Neumann memory-compute separation entirely. Each computing element stores its own state, removing the memory interface bottleneck. The architecture uses three physical dimensions via die stacking plus time as a fourth dimension, producing a chip that generates images at roughly 500 nanojoules versus millijoules on a GPU.
- •Sparsity as a Performance Multiplier: Reducing connections between computing elements — called sparsity — does not degrade system performance; it improves it. Sparse dynamical systems become more trainable and more scalable simultaneously, avoiding the n-squared connection scaling problem. This allows the architecture to grow without proportional energy or complexity costs, a combination previously considered unachievable.
- •Jevons Paradox and Market Scale: Dropping the cost of AI compute by 1,000x will not reduce total consumption — it will trigger demand expansion far exceeding the cost reduction, per Jevons Paradox. Rao projects this creates the largest market in human history, shifting infrastructure from centralized gigawatt data centers toward distributed, smaller facilities enabling billions of robotic deployments.
Notable Moment
Rao revealed publicly for the first time that Unconventional AI taped out its first physical dynamical computing chip on June 1 — just five months after the company formally launched — and has already produced real image outputs from it, demonstrating the architecture works in silicon.
Episode Transcript
Naveen Rao, cofounder and CEO of Unconventional AI, which is an AI chip startup. Best known for building and selling two deep tech companies. Naveen is kind of definitionally outlier founder. When I came there, we had about a $20,000,000 business and it was, you know, 7 or 800,000,000 when I left. I don't think you really understand something until you can build it. Just because something is tried does not mean it's wrong. I'm the opposite of an AI doomer. I think AI is the next evolution of humanity. We need innovation on hardware substrate to actually build true intelligence. Please welcome Naveen Rao. Hey, everyone. Great to be here. You know, switching gears a little bit to AI now, which you may have heard a little bit about. It's super exciting to be at this conference specifically because, as was said in the intro, I'm the opposite of a doomer. I think AI is one of the most transformational technologies that humanity's ever created, and will enable us to get to that next level of evolution, which I'm here for. And this is sort of the anti doomer conference, so let's go. So before we get going, I'll tell you a little bit about myself. You know, I It's kinda weird. I'm really right where I wanted to be my whole life. This was me at about five or six years old, something like that. We had a computer very early on, so I'll date myself, but this was in 1978 we got a computer. This is probably in the early eighties. Learned to program when I was a little kid. I just thought it was like a puzzle. You know, learned to I became an electrical engineer, really because I enjoyed sci fi, and always wanted to think about how I could make an intelligent machine. And, you know, then after a career in building computers, I went back to school and got a PhD in neuroscience. And the idea was like, let's go back to that thing. How do we make computers intelligent? And fortunately, the whole world kind of moved in this direction. So, you know, as a technologist, it's sort of the dream right now. A little bit about me tech from a company entrepreneurship standpoint, like, I actually founded the first AI chip company called Nirvana Systems. So this was in 2014. If anyone remembers back then, there was no AI, or at least not in the common vernacular. And it was really hard to actually convince people this is important, much less to build hardware around it. Now, you heard from Jensen up here, like, the largest company in the world is a hardware company because of AI. So we were early on. I think I sold the company way too early to Intel, but I I ran I started and ran the AI group at Intel. After I was done with that in 2020, I actually started thinking about the next …
Get the full transcript (4,503 words) + summary by email — free
One-time email with the complete transcript and AI summary of this episode. No account needed.
One email, no spam. We’ll also show you what SignalCast does.
Browse all All-In with Chamath, Jason, Sacks & Friedberg transcripts →
You just read a 3-minute summary of a 19-minute episode.
Get All-In with Chamath, Jason, Sacks & Friedberg summarized like this every Monday — plus up to 2 more podcasts, free.
Pick Your Podcasts — FreeKeep Reading
More from All-In with Chamath, Jason, Sacks & Friedberg
Steve Hilton & Spencer Pratt: Fixing California, Cheaper Gas, Ballot Fraud & CA Republicans
Sep 23 · 55 min
a16z Podcast
The 80-Year Bet: Why Naveen Rao Is Rebuilding the Computer from Scratch
Dec 8
More from All-In with Chamath, Jason, Sacks & Friedberg
Blake Scholl: Why Plane Speed Stalled, Supersonic Commercial Flight, & Revolutionizing the Engine
Sep 22 · 17 min
10% Happier with Dan Harris
The Science Of Emotion Regulation: A Five-Step Plan For Not Losing Your Sh*t | Marc Brackett
Sep 21
More from All-In with Chamath, Jason, Sacks & Friedberg
We summarize every new episode. Want them in your inbox?
Steve Hilton & Spencer Pratt: Fixing California, Cheaper Gas, Ballot Fraud & CA Republicans
Blake Scholl: Why Plane Speed Stalled, Supersonic Commercial Flight, & Revolutionizing the Engine
Adam Foroughi, Applovin CEO: Surviving a 92% Drawdown, Ads as ML 1.0 & the $50B Game Ad Market
Bill Gurley: Searching for Feynman
Jared Isaacman: A New Era for NASA and American Space Exploration
Similar Episodes
Related episodes from other podcasts
a16z Podcast
Dec 8
The 80-Year Bet: Why Naveen Rao Is Rebuilding the Computer from Scratch
10% Happier with Dan Harris
Sep 21
The Science Of Emotion Regulation: A Five-Step Plan For Not Losing Your Sh*t | Marc Brackett
Deep Questions with Cal Newport
Aug 10
How Do I Finish Meaningful Projects? | Monday Advice
This Week in Startups
Jun 1
This Startup Fused Human Brain Cells with Silicon Chips | E2295
NVIDIA AI Podcast
Apr 14
How AI Will Change Quantum Computing - Ep. 294
Explore Related Topics
This podcast is featured in Best Tech Podcasts (2026) — ranked and reviewed with AI summaries.
Read this week's Startups & Product Podcast Insights — cross-podcast analysis updated weekly.
You're clearly into All-In with Chamath, Jason, Sacks & Friedberg.
Every Monday, we deliver AI summaries of the latest episodes from All-In with Chamath, Jason, Sacks & Friedberg and 192+ other podcasts. Free for one show.
Start My Monday DigestNo credit card · Unsubscribe anytime