Physical AI Startup Ethernovia Raises $90M as Investors Bet Big on Real-World AI
What is “physical AI,” and why is it suddenly attracting nine-figure investments? As artificial intelligence moves beyond chatbots and recommendation engines, a new wave of innovation is taking shape—AI that interacts with the physical world through robotics, smart infrastructure, and autonomous vehicles. At the heart of this shift lies Ethernovia, a San Jose–based semiconductor startup that just raised $90 million in Series B funding to accelerate its mission: building the data highways that connect sensors to decision-making brains in real time.
Unlike traditional AI models that operate purely in the digital realm, physical AI demands ultra-low latency, high-bandwidth communication between dozens—or even hundreds—of sensors and central processing units. That’s where Ethernovia comes in. Its custom Ethernet-based processors are designed to move massive streams of sensor data across complex systems without bottlenecks, a critical capability for everything from self-driving cars to warehouse robots.
Why “Physical AI” Is the Next Frontier for Tech Investors
The term “physical AI” has gained traction in 2025 and 2026 as investors recognize a fundamental truth: the most transformative applications of artificial intelligence won’t live on screens—they’ll operate in factories, on roads, and in homes. Think robotic arms assembling EVs with millimeter precision, delivery drones navigating urban canyons, or surgical robots assisting in operating rooms. All of these require AI that not only “thinks” but also acts—and acts fast.
This shift has redirected venture capital toward companies enabling the hardware-software bridge. While headlines often spotlight flashy AI models, the real bottleneck lies in data movement. A single autonomous vehicle can generate over 1 terabyte of sensor data per hour. Without efficient, reliable interconnects, that data becomes useless—or worse, dangerous.
Ethernovia’s technology addresses this by reimagining Ethernet—not just as a networking standard, but as a real-time nervous system for intelligent machines.
Inside Ethernovia’s High-Speed Data Architecture
Founded in 2021, Ethernovia has quietly built a reputation among automotive and industrial tech insiders for its deterministic Ethernet solutions. Unlike conventional Ethernet, which prioritizes throughput over timing, Ethernovia’s processors guarantee precise data delivery within microseconds—a non-negotiable requirement for safety-critical systems.
The company’s chips integrate time-sensitive networking (TSN) protocols directly into silicon, enabling synchronized communication across lidar, radar, cameras, and inertial measurement units. This synchronization ensures that an autonomous vehicle doesn’t “see” a pedestrian in one frame and misplace them in the next due to network jitter.
“We’re not just moving data—we’re preserving the integrity of time and space across distributed systems,” said a company spokesperson. “In physical AI, milliseconds aren’t just delays—they’re margins of safety.”
Already, Ethernovia’s processors are being evaluated by major automakers and Tier 1 suppliers, with pilot deployments expected in late 2026.
Maverick Capital’s Bold Bet on the AI Hardware Stack
The $90 million round was led by Maverick Silicon, a specialized AI fund launched in 2024 by the storied hedge fund Maverick Capital. Notably, this marks Maverick’s first-ever sector-specific investment vehicle in its 30-year history—a signal of how seriously the firm views the physical AI opportunity.
“Generative AI captured the imagination, but physical AI will capture the economy,” said a Maverick Silicon partner in a statement. “The companies that win won’t just have great algorithms—they’ll own the infrastructure that makes real-world AI possible.”
Existing investors Porsche SE and Qualcomm Ventures also participated, underscoring cross-industry confidence. Porsche’s involvement hints at potential integration into future high-performance electric vehicles, while Qualcomm’s backing suggests synergies with its Snapdragon Ride autonomous driving platform.
The Quiet Infrastructure Boom Behind the AI Hype
While consumers see AI through the lens of chatbots and image generators, a parallel revolution is unfolding in industrial corridors and R&D labs. According to PitchBook, investments in “AI enablers”—companies providing chips, sensors, and connectivity solutions for physical systems—surged 170% in 2025 alone.
Ethernovia sits squarely in this category. It doesn’t train large language models; it ensures that when a robot arm receives a command, it executes it with zero lag. In an era where AI reliability is paramount, such unglamorous but essential components are becoming strategic assets.
Analysts note that as regulatory bodies like the NHTSA and EU Commission tighten safety standards for autonomous systems, demand for deterministic, certifiable hardware will only grow. Ethernovia’s focus on automotive-grade reliability positions it well for this shift.
What This Means for the Future of Autonomous Systems
The implications of Ethernovia’s technology extend far beyond cars. Industrial automation, smart cities, and even defense applications all rely on tightly coordinated sensor networks. A warehouse robot fleet using Ethernovia’s chips could share real-time spatial awareness, avoiding collisions while optimizing logistics routes. In agriculture, autonomous tractors could synchronize soil and crop data across fields with unprecedented accuracy.
Critically, this infrastructure must be scalable and cost-effective. Ethernovia claims its solution leverages standard Ethernet protocols, avoiding proprietary lock-in and enabling easier integration into existing manufacturing ecosystems—a key advantage over competitors relying on custom interconnects.
As physical AI scales from prototypes to mass deployment, interoperability and supply chain resilience will be as important as raw performance. Here, Ethernovia’s use of widely adopted Ethernet standards gives it a strategic edge.
A New Chapter in the AI Investment Cycle
The Ethernovia funding round reflects a maturing AI investment landscape. After the initial frenzy around foundational models, capital is now flowing downstream—to the layers that make AI actionable in the real world. This “stack diversification” mirrors earlier tech cycles, such as the shift from internet portals to cloud infrastructure in the 2000s.
For startups, the message is clear: solving hard engineering problems in data movement, power efficiency, and real-time control is now as valuable as algorithmic breakthroughs. For enterprises, it means partnerships with enablers like Ethernovia will be critical to deploying safe, scalable physical AI systems.
As one industry insider put it: “We’ve spent years teaching machines to think. Now we need to teach them to move—and move together.”
The Invisible Engines of the AI Age
Ethernovia may not have a consumer brand, but its chips could soon be embedded in millions of intelligent machines worldwide. In the race to build physical AI, the winners won’t always be the ones with the flashiest demos—they’ll be the ones ensuring data flows flawlessly, reliably, and instantly across the physical-digital divide.
With $90 million in fresh capital and heavyweight backers betting on its vision, Ethernovia is poised to become a quiet powerhouse in the AI infrastructure stack. And as autonomous systems inch closer to everyday reality, the world may soon depend on the very technology this unassuming startup is perfecting—one microsecond at a time.
