Earlier this week, five individuals spanning the entire AI supply chain gathered at the Milken Global Conference in Beverly Hills. They discussed everything from chip shortages to orbital data centers—and even the possibility that the fundamental architecture powering today’s AI might be flawed. (Yan LeCun, Meta’s former chief AI scientist, became the founding chair of its technical research board earlier this year.)

Here’s what they had to say.

**The Bottlenecks Are Real**

The AI boom is hitting hard physical limits, and the constraints start deeper in the stack than many realize. Fouquet kicked things off, describing a “huge acceleration in chip manufacturing” but expressing his “strong belief” that despite all that effort, “for the next two, three, maybe five years, the market will be supply limited.” That means hyperscalers like Google, Microsoft, Amazon, and Meta won’t get all the chips they’re paying for—full stop. DeSouza highlighted just how massive and fast-growing this issue is, noting that Google Cloud’s revenue crossed $20 billion last quarter (up 63%), and its backlog—committed but not yet delivered revenue—nearly doubled in a single quarter, from $250 billion to $460 billion. “The demand is real,” he said with remarkable calm. For Younis, the constraint comes from elsewhere. Applied Intuition builds autonomy systems for cars, trucks, drones, mining equipment, and defense vehicles. His bottleneck isn’t silicon—it’s data that can only be gathered by sending machines into the real world and observing what happens. “You have to find it from the real world,” he said. No amount of synthetic simulation fully closes that gap. “There will be a long time before you can fully train models that run on the physical world synthetically.”

**The Energy Problem Is Also Real**

If chips are the first bottleneck, energy is the one looming behind. DeSouza confirmed that Google is exploring data centers in space as a serious response to energy constraints. “You get access to more abundant energy,” he noted. But even in orbit, it’s not simple. Space is a vacuum, so convection is eliminated, leaving radiation as the only way to shed heat into the surrounding environment—a much slower and harder-to-engineer process than the air and liquid cooling systems data centers rely on today. Still, Google treats it as a legitimate path. DeSouza’s deeper argument, unsurprisingly, was about efficiency through integration. Google’s strategy of co-engineering its full AI stack—from custom TPU chips through to models and agents—pays dividends in watts per flop that a company buying off-the-shelf components simply can’t replicate, he suggested. “Running Gemini on TPUs is much more energy efficient than any other configuration,” because chip designers know what’s coming in the model before it ships. In a world where energy availability is becoming a massive constraint, that kind of vertical integration is a major competitive advantage. Fouquet echoed the point later: “Nothing can be priceless.” The industry is in a strange moment, investing extraordinary capital driven by strategic necessity. But more compute means more energy, and more energy has a price.

**A Different Kind of Intelligence**

While the rest of the industry debates scale, architecture, and inference efficiency within the large language model paradigm, Bodnia is building something very different. Her company, Logical Intelligence, is based on so-called energy-based models (EBMs)—a class of AI that doesn’t predict the next token in a sequence but instead tries to understand the rules underlying data. She argues this is closer to how the human brain actually works. “Language is a user interface between my brain and yours,” she said. “The reasoning itself is not attached to any language.”

Her largest model runs at 200 million parameters—compared to the hundreds of billions in leading LLMs—and she claims it runs thousands of times faster. More importantly, it’s designed to update its knowledge as data changes, rather than requiring retraining from scratch. For chip design, robotics, and other domains where a system needs to grasp physical rules rather than linguistic patterns, she argues EBMs are a more natural fit. “When you drive a car, you’re not searching for patterns in any language. You look around you, understand the rules about the world around you, and make a decision.” It’s an intriguing argument, and one likely to attract more attention in the coming months, as the AI field begins to question whether scale alone is sufficient.

**Agents, Guardrails, and Trust**

Shevelenko spent much of the conversation explaining how Perplexity has evolved from a search product into something it now calls a “digital worker.” Perplexity Computer, its newest offering, is designed not as a tool a knowledge worker uses, but as staff that a knowledge worker directs. “Every day you wake up and you have a hundred staff on your team,” he said of the opportunity. “What are you going to do to make the most of it?”

It’s a compelling pitch—but it raises obvious questions about control. When asked, his answer was granularity. Enterprise administrators can specify not just which connectors and tools an agent can access, but whether those permissions are read-only or read-write—a distinction that matters enormously when agents act inside corporate systems. When Comet, Perplexity’s computer-use agent, takes actions on a user’s behalf, it presents a plan and asks for approval first. Some users find the friction annoying, Shevelenko said, but he considers it essential, especially after joining the board of Lazard, where he found himself unexpectedly sympathetic to the conservative instincts of a CISO protecting a 180-year-old brand built entirely on client trust. “Granularity is the bedrock of good security hygiene,” he said.

**Sovereignty, Not Just Safety**

Younis offered what may have been the panel’s most geopolitically charged observation: physical AI and national sovereignty are entangled in ways purely digital AI never was. The internet initially spread as American technology and faced pushback only at the application layer—the Ubers and DoorDashes—when offline consequences became visible. Physical AI is different. Autonomous vehicles, defense drones, mining equipment, and agricultural machines manifest in the real world in ways governments can’t ignore, raising questions about safety, data collection, and who ultimately controls systems that operate inside a nation’s borders. “Almost consistently, every country is saying: we don’t want this intelligence in a physical form in our borders, controlled by another country.” Fewer nations, he told the crowd, can currently field a robotaxi than possess nuclear weapons.

Fouquet framed it differently. China’s AI progress is real—DeepSeek’s release earlier this year sent something close to a panic through parts of the industry—but that progress is constrained below the model layer. Without access to EUV lithography, Chinese chipmakers cannot manufacture the most advanced semiconductors, and models built on older hardware operate at a compounding disadvantage no matter how good the software gets. “Today, in the United States, you have the data, you have the computing access, you have the chips, you have the talent. China does a very good job on the top of the stack, but is lacking some elements below,” Fouquet said.

**The Generation Question**

Near the end of the panel, someone in the audience asked the obvious uncomfortable question: will all of this impact the next generation’s capacity for critical thinking? The answers were, perhaps unsurprisingly, optimistic—though not naively so. DeSouza pointed to the scale of problems that more powerful tools might finally let humanity address: neurological diseases whose biological mechanisms we don’t yet understand, greenhouse gas removal, and grid infrastructure that has been deferred for decades. “This should unleash us to the next level of creativity,” he said.

Shevelenko made a more pragmatic point: the entry-level job may be disappearing, but the ability to launch something independently has never been more accessible. “For anybody who has Perplexity Computer, the constraint is your own curiosity and agency.”

Younis drew the sharpest distinction between knowledge work and physical labor. He noted that the average American farmer is 58 years old, and labor shortages in mining, long-haul trucking, and agriculture are chronic and growing—not because wages are too low, but because people don’t want those jobs. In those domains, physical AI isn’t displacing willing workers. It’s filling a void that already exists and looks only to deepen from here.