Why it might not make sense for you to own a self-driving car

2026-05-14 · Source: Understanding AI · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning · Depth: Fundamental Awareness, short

Summary

A prototype self-driving car from the startup Tensor, formerly AutoX, was recently showcased in San Francisco, revealing its ambitious plan to sell fully autonomous vehicles directly to consumers. Tensor, founded in 2016, previously tested robotaxi services but rebranded last year to focus on ownership. The vehicle features a Waymo-like sensor rack, internal video screens replacing side mirrors, and a retractable steering wheel. It boasts eight Nvidia Thor GPUs, delivering 8,000 trillion operations per second (TOPS), significantly exceeding Tesla's Hardware 3 (144 TOPS) and projected AI 5 (2,500 TOPS) chips. With each Nvidia Thor chip costing around \$3,499, the car's computing power alone is expensive. Tensor's chief marketing officer, Amy Luca, described the car as "luxury," with prices potentially exceeding Waymo's rumored \$150,000. The design includes a large water tank for months-long sensor cleaning and will require mandatory maintenance and monthly service fees, with a planned launch in the UAE this year and the US next year.

Key takeaway

For investors or entrepreneurs evaluating the consumer self-driving car market, understand that the path to profitability is fraught with high hardware costs and ongoing service demands. Your projections must account for vehicles requiring 8,000 TOPS computing power and luxury price points, alongside mandatory maintenance fees. This model suggests a niche, high-end market rather than mass adoption, impacting your scaling strategies and return on investment.

Key insights

Consumer ownership of fully self-driving cars faces significant hurdles due to extreme hardware costs and complex maintenance requirements.

Principles

• Unsupervised self-driving demands extreme computing power.
• Sensor maintenance is a critical, ongoing ownership burden.
• Liability concerns drive mandatory service agreements.

In practice

• Consider total cost of ownership for advanced vehicle tech.
• Evaluate computing power needs for autonomous systems.
• Factor in mandatory service contracts for liability-sensitive products.

Topics

• Autonomous Vehicles
• Self-Driving Car Ownership
• Tensor
• NVIDIA Thor
• Vehicle Computing Power
• Sensor Maintenance

Best for: Computer Vision Engineer, AI Product Manager, Entrepreneur, Investor

Related on AIssential

• Elon Musk admits millions of Tesla owners need upgrades for true ‘Full Self-Driving’ — Tesla's Hardware 3 is insufficient for unsupervised Full Self-Driving, necessitating extensive upgrades.
• The Data Moat for Self Driving — Extensive real-world driving data, capital, and GPU infrastructure form an insurmountable moat for self-driving development.
• Sector Snapshot: Autonomous Vehicle Funding More Than Triples In 2026 To Hit Record Amount — Autonomous vehicle investment is consolidating into a few major players ready for commercial scaling.
• Tesla hits Musk’s threshold for ‘safe unsupervised’ driving — Tesla's FSD (Supervised) has logged 10 billion miles, yet unsupervised driving for customers remains elusive due to liability concerns.
• Meta to Deploy Homegrown Chips, Uber to Offer Zoox Rides — AI demand drives significant investment in computing infrastructure, custom silicon, and autonomous systems across multiple industries.
• Self-driving scaleup Wayve raises fresh funds from AMD, Qualcomm and Arm — Wayve secured $60M from chipmakers to scale its AI self-driving software across diverse automotive compute platforms.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Understanding AI.