BrightAI Raises $97M to Push Physical AI Into Power Grids and Water Systems
BrightAI raised $97M led by Smith Point Capital, with GE Vernova joining, to scale its edge-first Physical AI platform after finding 98% of 12,158 operators lack asset visibility.
By Daniel Okafor
3 min read

What's News
- BrightAI raised $97M on November 18, 2026, led by Smith Point Capital, bringing total funding to $184M.
- BrightAI reports 98% of 12,158 measured operators across 16 industries lacked adequate asset visibility; estimated addressable annual impact exceeds $7.5 trillion.
- Stateful runs across 50,000+ locations, integrating millions of sensors and processing tens of trillions of readings.
- Customer GE Vernova joined the round and will deploy the tools across its Power, Wind and Electrification segments.
- Nearly 60% of the addressable opportunity sits outside North America, in Asia and Europe.
BrightAI has raised $97 million to scale a Physical AI platform that watches power plants, water networks and pipelines from the sensor to the field worker. Smith Point Capital led the round, announced November 18, 2026. The deal lifts BrightAI's total capital raised to $184 million.
Cobalt Capital, Marlinspike Partners, Zero Infinity Partners and existing customer GE Vernova joined the financing. Earlier investors Khosla Ventures, Inspired Capital and Upfront Ventures also participated. GE Vernova was a customer first. It will deploy BrightAI's tools across its Power, Wind and Electrification segments to augment field-inspection expertise, according to the announcement.
Why $97 million now?
The company's pitch rests on one number: 98% of the 12,158 operators BrightAI measured across 16 essential-service industries lacked adequate visibility into the condition of the assets they run or oversee. BrightAI estimates the addressable annual impact at more than $7.5 trillion worldwide.
Energy, water, waste and heavy-industry operators still work on an inspect, detect and react model. More than half of scheduled inspections find nothing wrong, according to BrightAI, while consequential failures can stay unseen until they are underway. Crews follow calendars rather than current asset condition.
Smith Point's thesis explains the product design. Cloud-only software is weakest where infrastructure sits underground, on poles or across remote sites with scarce power, connectivity and human presence — exactly where BrightAI's edge-first stack does its work.
How does the Stateful platform work?
BrightAI's platform, called Stateful, connects edge sensors, autonomous machines and wearable copilots to a shared foundation-model layer. It monitors asset conditions, identifies issues, supplies step-by-step guidance and feeds field outcomes back into its models. BrightAI builds the sensors, edge compute, models and field systems as one stack, so operational data never passes through another platform and remains the operator's asset.
The core hardware is small. The Stateful Sticker captures vibration, temperature and acoustic signals, with visual monitoring available depending on device configuration. Foundation models run on the device itself. They can flag anomalies in real time even when the network drops, and low-power wireless communication removes the need for a continuous internet connection.
For closer work, the Autonomous Inspection layer integrates with compatible drones and robots. These machines assess utility poles, pipelines, gauges and valves through visual and thermal analysis, cutting the need to send personnel into hazardous or hard-to-reach locations. BrightAI reports its autonomous systems deliver more than five times the productivity of manual inspection.
Stateful Wearable closes the loop. Technicians get voice and text guidance for procedures and troubleshooting, the device captures field observations, and completed activities become structured work records. A sensor reading triggers a machine inspection, the inspection triggers guided work, and the finished job adds data back to the platform.
Where is it already running?
Stateful runs across more than 50,000 operational locations, according to BrightAI. The platform integrates millions of sensors and connected systems and has processed tens of trillions of real-world sensor readings. It monitors hundreds of thousands of critical assets and guides tens of thousands of frontline workers and autonomous systems.
Customers and co-development partners include GE Vernova, Reworld, Osmose Utilities Services, Azuria Water Solutions, Kodiak Gas Services and Diamondback Energy.
What comes next?
Nearly 60% of BrightAI's addressable opportunity sits outside North America, concentrated in Asia and Europe, the company says. It is opening offices and field operations in key international markets. The groundwork is already domestic: a 25,000-square-foot Palo Alto headquarters, an operations center in Midland, Texas, a briefing center in Washington, D.C., and a headcount that doubled in 2025.
The financing ties an international push to a broader GE Vernova deployment and a product stack that learns from sensors, inspections and field work. The execution question is whether that footprint can deliver earlier, more reliable failure detection across essential services — while keeping operational data under operators' control.
Source: unite.ai
More from Daniel Okafor
Show full bio
Correspondent covering money, funding and market shifts around entrepreneurship, startups and venture funding at Business Bearings.
2,360 articles