Written by Cassian Vance
The software-centric artificial intelligence boom that dominated the early 2020s is officially transitioning into its next evolutionary phase: Physical AI. For investors, the shift from large language models processing text to robotic foundation models interacting with the physical world represents the next major thematic supercycle. The transition from bits to atoms is no longer a theoretical exercise; it is an investable reality supported by massive capital deployment, breakthrough foundation models, and accelerating commercial adoption.
The market dynamics are compelling. Grand View Research projects the artificial intelligence in robotics market will expand from $20.4 billion in 2025 to $182.7 billion by 2033, representing a staggering 32% compound annual growth rate. This exponential trajectory is driven by a fundamental shift in how machines learn. Instead of being rigidly programmed for specific tasks, modern robotic systems are leveraging imitation learning and generalist foundation models to acquire skills autonomously. The investment thesis is clear: the companies providing the silicon, simulation environments, and hardware for Physical AI will capture outsized returns in the coming decade.
The private markets are already signaling the momentum behind this shift. On May 19, Axis Robotics (axisrobotics.ai), a data-to-model infrastructure provider, secured a $10 million funding round led by Hack VC. The company is building the critical scaling layer for Physical AI, transforming human teleoperation into deployable robotic intelligence. Their platform has already aggregated over one million data trajectories in a single month from 80,000 global contributors, achieving a 90% success rate on sim-to-real tasks. This flow of venture capital into foundational infrastructure companies like Axis Robotics—founded by researchers from UC Berkeley, CMU, UCLA, and NTU—underscores the institutional conviction that Physical AI is the next major platform shift.
For public market investors, gaining exposure requires a nuanced approach, separating the foundational infrastructure providers from the hardware manufacturers. The following analysis evaluates three leading publicly traded companies positioned to benefit from the Physical AI supercycle.
NVIDIA Corporation (NASDAQ: NVDA)
NVIDIA remains the undisputed foundational layer for the Physical AI revolution. While the market has largely priced the company based on its data center GPU dominance for large language models, its comprehensive robotics stack—spanning the Omniverse simulation platform, Isaac robotics framework, and the new GR00T foundation models for humanoid robots—represents a massive, underappreciated growth vector.
The company’s strategy is to provide the end-to-end ecosystem required to train, simulate, and deploy Physical AI. By creating digital twins in Omniverse, robotics companies can train models in physically accurate virtual environments before deploying them in the real world. The recent introduction of the Isaac GR00T N1.7 model, which allows humanoid robots to take natural language instructions, demonstrates NVIDIA’s lead in bridging the gap between cognitive AI and physical execution.
NVIDIA’s stock is currently trading near $220, having pulled back slightly from recent highs. Despite this, the company’s projected 2026 earnings power of over $190 billion suggests the valuation remains reasonable given its monopoly-like position in AI infrastructure. The transition from software AI to Physical AI will require an order of magnitude more compute for simulation and real-world inference, extending NVIDIA’s growth runway well into the 2030s.
Verdict: BUY
Price Target: $320 (Representing a 45% upside from current levels, aligned with Bank of America’s recent estimates, driven by the expanding total addressable market for robotic simulation and inference.)
Tesla, Inc. (NASDAQ: TSLA)
Tesla represents the most ambitious, yet highest-risk, pure-play on consumer and industrial humanoid robotics through its Optimus program. CEO Elon Musk has consistently positioned Tesla not as an automotive manufacturer, but as an AI and robotics company. The company’s unique advantage lies in its massive real-world data collection engine derived from its Full Self-Driving (FSD) fleet, which provides a significant edge in training vision-based neural networks for physical navigation.
The timeline for Optimus commercialization remains fluid. While the company missed its aggressive 2025 targets, recent progress on the Gen 3 hardware and the planned integration into the new Texas manufacturing facility indicate that mass production is moving closer to reality. The long-term vision is staggering: targeting 10 million units with a cost structure under $20,000 per robot. If executed, this would fundamentally disrupt global labor markets.
However, the current valuation, with the stock trading around $405, already bakes in significant optimism regarding the successful execution of both the robotaxi and Optimus initiatives. The automotive business is facing margin compression and slowing growth, as evidenced by the recent Q1 2026 results where automotive gross margins expanded only slightly while energy revenues declined. Investors buying Tesla today are paying a premium for a robotics future that remains several years away from meaningful revenue contribution.
Verdict: HOLD
Price Target: $420 (Reflecting the near-term headwinds in the core automotive business balanced against the immense long-term optionality of the Optimus program. Investors should wait for a more attractive entry point or clearer evidence of commercial scaling for Optimus.)
FANUC Corporation (OTC: FANUY)
For investors seeking immediate, profitable exposure to industrial Physical AI without the speculative premium attached to consumer humanoids, Japan-based FANUC offers a compelling value proposition. As the global leader in factory automation and industrial robotics, FANUC possesses the installed base and manufacturing expertise that Silicon Valley software companies lack.
FANUC has astutely recognized that the future of industrial automation requires partnering with AI leaders rather than attempting to build foundation models internally. In May 2026, the company announced a landmark partnership with NVIDIA to integrate the Isaac GR00T foundation model and Omniverse simulation into its massive fleet of over 200 types of industrial robots. Simultaneously, FANUC partnered with Google to bring Gemini AI and Intrinsic software to its systems, creating a robust, dual-platform AI stack.
This strategic positioning allows FANUC to upgrade its existing hardware install base with cutting-edge Physical AI capabilities, transforming rigid, pre-programmed robots into adaptive systems capable of learning tasks like complex assembly and manipulation through imitation. Trading around $25 on the OTC markets (up nearly 24% year-to-date), FANUC offers a reasonable valuation for a company that is actively shipping revenue-generating products and holds a massive backlog, insulating it from the speculative volatility of pure-play AI stocks.
Verdict: BUY
Price Target: $32 (Based on the anticipated margin expansion and service revenue growth derived from integrating advanced AI software capabilities into its dominant industrial hardware footprint.)
The Road Ahead
The transition to Physical AI is the most significant technological shift since the advent of the smartphone. While software AI demonstrated the power of cognitive reasoning, Physical AI will automate the execution of tasks in the real world, unlocking trillions of dollars in economic value. Investors must position their portfolios now, focusing on the infrastructure providers supplying the compute and simulation environments, and the established hardware manufacturers successfully integrating these new capabilities. The funding flowing into private companies like Axis Robotics confirms the institutional thesis: the future of AI is physical, and the investment supercycle is just beginning.
Disclaimer: This article is for informational purposes only and does not constitute investment advice. The author may hold positions in securities mentioned. Past performance does not guarantee future results.
