Ph Physical AI Loading... : Investor Sentiment and Bull/Bear Views

AI has moved from learning to inference and toward AGI, but the ultimate endpoint is physical AI. When AI-embedded mobility or autonomous vehicles become visibly commercialized, the market can re-rate the physical AI theme.

Ian is very bullish on physical AI and robotics, viewing AI's entry into the physical world as potentially as transformative as LLMs. Cantos invests across the robotics stack, from model layer to dexterity and actuation, because movement is a huge category alongside energy and information.

Katelin sees the physical AI and hardware boom as more than a bubble: the best engineers have shifted from ad tech to building rocket engines and hard tech, and hardware is now treated as a moat. She echoes Jensen's prediction that every industrial company will become a robotics company.

With language models largely dominated by frontier players, the next massive investment opportunities lie in applying transformer AI to physical robotics and biology, spaces that are currently heating up.

Kalanick argues that the next industrial revolution is the digitization of the physical world: software, sensors, robotics, and AI will automate entire industrial sectors. He calls this industrial AI and frames it as a massive opportunity because it targets multiple hundred-billion- to trillion-dollar industries, unlike digital AI, which mainly affects the enterprise.

Instead of overvalued AI infrastructure, the better opportunity lies in physical AI, robotics, and automation, which will benefit as the underlying infrastructure is built out.

Andrew Kang believes robotics/physical AI is at a pre-ChatGPT inflection point and is underappreciated by investors. He points to early robot foundation models (Google DeepMind's RT-1/RT-2) showing generalizability, which can extend to simple factory tasks. He expects tens of billions of industry revenue within 2-4 years and sees a multi-decade, multi-trillion-dollar opportunity if humanoids reach human-like physical labor economics. He is focusing all his time and capital here.

Physical AI—putting intelligence into cars, trucks, tanks, drones, manufacturing, mining, logistics, and supply chains—will transform the global economy; the companies that impact the physical world may ultimately be bigger than the digital AI winners. Applied Intuition's mission is to put intelligence on a billion machines.

AI can learn any structured domain—proteins, genes, cells, human anatomy, physics, climate, 3D worlds, robotics, and self-driving cars. Jensen says the physical part of the economy is about $80 trillion and is the most important frontier, beyond just language-model chatbots.

Cameron expects all hardware companies to become physical AI companies, with AI agents optimizing design, manufacturing, testing, and field monitoring. Because physical products must be rigorously validated in the real world, he sees an explosion in hardware testing and a race to collect, clean, and fuse real-world test data with model outputs.

CES 2026 marked a humanoid-robot inflection from showmanship to use cases, with 2026 becoming a proof-of-concept war; robots will enter dangerous, dull and repetitive industrial jobs first, and caregiving has large ROI potential as labor supply shrinks.

CES 2026 marked an inflection point for humanoid robots: the industry is shifting from performance demonstrations to commercialization, with companies that clearly define paid use cases and ROI likely to lead. The key question is no longer what a robot can do, but how it will be deployed and monetized.

CES showed physical AI arriving through robotics and autonomous systems; this fourth industrial revolution is bullish for robotics.

She calls physical AI the next big thing and expects robotics to be a major part of it, along with industrial applications where edge AI comes into play. AMD is tuning its technology and capabilities for these evolving markets, which she views as a natural extension of AI.

The enabling technology for physical AI and robotics is now ready because generative models can produce action tokens, such as steering-wheel activation or finger articulation, just as they generate language or video. Jensen sees this as a long-pursued moment whose time has come, with autonomous vehicles and humanoid robots as examples.

Choi makes humanoid and physical AI his top 2026 CES and HORSE theme. The 2026 CES slogan is Innovator Show Up, and he says humanoid robots must prove real-world capabilities beyond dancing or taekwondo. Physical AI, meaning AI entering the real economy, will be the key theme, and CES will showcase many humanoid and robotics companies.

CES 2026's first key theme is physical AI, covering robots, autonomous vehicles, and all moving machines. The long-term opportunity is large, but current robots mainly imitate learned motions and lack true learning or judgment, so progress in sensors and autonomous driving is needed.

Ives expects physical AI, humanoid robotics, and autonomous systems to be front and center in 2026, with embedded AI devices gaining prominence. He identifies Nvidia and Tesla as the two best physical AI plays.

Haas agrees physical AI will be a gigantic chip market, potentially bigger than data centers on a unit basis. Robots today mostly use repurposed automotive chips built for ADAS functional-safety compliance, which are not designed for actuators or the smaller parts of a joint, so dedicated silicon is needed; each robot will carry tens to hundreds of chips, and on-device AI that can learn makes the unit volume far larger than anything in the market today.

Su is a big believer in physical AI but deliberately puts a timeline on it: the market for physical-AI chips will take at least five years, and she says five plus, to become larger than the market for data-center chips. She expects it to become a significant end market rather than an imminent one, with chips in data centers and chips at the edge both remaining significant markets in the meantime. The practical read is to monitor the robotics and edge-compute opportunity while data-center demand remains the dominant driver.

Huang says everything in the world that moves will be autonomous someday and that someday is probably around the corner. The structural consequence is that every company that builds machines will run two factories - the machine factory and an AI factory that produces the brain for those machines - and he expects the same pattern to spread into areas such as air traffic control where a giant AI does the monitoring and a person intercepts only in exceptions. His conclusion is that every industrial company in the future will be an AI company or it will not be an industrial company, making autonomy a broad investable regime rather than a niche.