KIDZ AI Inc., an education technology company specializing in AI infrastructure, GPU computing and physical AI, is pondering whether to add another offering to its growing list: operating a fleet of Tesla Robotaxis.

Its consideration of acquiring a fleet lies somewhere between a thinking-outside-the-box idea tossed around during a board meeting and actually raising the capital, buying the vehicles and putting them to work on city streets. For a company rooted in education technology, it would be quite a commute.

KIDZ AI announced Wednesday, Sept. 30, that it had submitted a fleet procurement inquiry to Tesla for an initial 500 Robotaxis, the first step in what the company is calling an Autonomous Fleet Operations Strategy. The idea is to acquire autonomous vehicles, deploy them commercially and manage them as revenue-producing assets.

KIDZ AI has not committed to buying 500 vehicles, and Tesla has not committed to selling or delivering them. The company described the proposal as being in the planning, evaluation and partner-exploration stage and said the inquiry is nonbinding. KIDZ AI could decide that operating robotaxis is too expensive, too complicated or simply not worth pursuing.

If the inquiry advances, the process would move into price negotiations, financing, vehicle availability, regulatory approvals, insurance, maintenance, charging, fleet management and determining where the cars could legally and profitably operate. KIDZ AI said it would weigh acquisition and operating costs against potential revenue before making a decision.

Instead of investing solely in the computing infrastructure that powers AI, KIDZ AI would own physical assets capable of using AI to generate revenue.

KIDZ AI develops learning systems combining AI, AI agents and robotics, while also pursuing GPU cloud computing and data-center infrastructure. Autonomous vehicles fit into the broader field often described as physical or embodied AI: AI that does not merely generate words or images but perceives its surroundings, makes decisions and acts in the physical world.

A robotaxi may represent a departure from KIDZ AI’s core business, but not necessarily from the technological direction it has been pursuing. And the company is leaving itself room to look beyond Tesla, saying it intends to evaluate other autonomous platforms and applications beyond passenger robotaxis as the market develops.

Morgan Stanley Research estimated in July that the global robotaxi market could reach roughly $1 trillion by 2040, encompassing mobility services, vehicles, software and related businesses. Its analysts forecast about 2.5 million robotaxis operating worldwide by 2035 and said the industry could reach break-even around 2028 as hardware and operating costs decline.

“The journey is just beginning and substantial hurdles remain, but we are at an inflection point,” Tim Hsiao, Morgan Stanley’s Greater China auto and shared mobility equity research analyst, said in the report.

Alphabet’s Waymo has built a large commercial driverless network, while Tesla is operating its Robotaxi service in limited areas of Austin, Dallas, Houston, Miami, Orlando and Tampa. Amazon-owned Zoox has begun commercial service with purpose-built vehicles that dispense with conventional driver controls altogether. Mobileye, known for supplying automated-driving technology to automakers, announced this year that it plans to launch its own U.S. robotaxi service in 2027.

NVIDIA provides the high-powered computing needed to process the enormous amounts of information autonomous systems collect. In China, Baidu’s Apollo Go has emerged as another major robotaxi operation.

Federal regulators continue to collect crash reports involving automated driving systems and advanced driver-assistance technology, and the National Highway Traffic Safety Administration said this summer that it had identified a pattern of driverless vehicles interfering with law enforcement officers and other first responders. Waymo, despite reporting substantially lower injury-crash rates than human drivers in comparable operating environments, has also faced recalls and investigations involving how its vehicles respond to unusual road conditions.

Those concerns have not stopped the money or expansion. Waymo raised $16 billion this year at a $126 billion valuation and has been pushing into additional markets. Zoox received federal approval in July for limited commercial deployment of robotaxis built without steering wheels or other traditional human controls.

For KIDZ AI, all of that makes 500 Robotaxis an intriguing proposition, but still only a proposition.