Wood Mackenzie: Why Autonomous EVs are Ready to Scale

The autonomous EV (AEV) industry could shift from experimental testing to mainstream commercial deployment in 2026 according to Wood Mackenzie.

The Edinburgh-based consultancy expects AEV operations or testing in 39 markets by the end of 2026.

This implies a significant acceleration in their rollout despite some existing economic and functional uncertainties around the technology.

In total, Wood Mackenzie outlines four major trends that it believes could come to define 2026 in the world of driverless vehicles. 

Developments in AI systems

The shift could hinge on new Vision-Language-Action AI models, or VLAs.

These models replace traditional "rule-based" systems and expensive high-definition mapping with camera-and-video perception technology that is able to make fast decisions in real time.

Thanks to recent improvements in this kind of technology, the expected timelines for the city launches of many AEV fleets have been shortened considerably.

"New AI technologies are fundamentally transforming the economics of deployment," says David Brown, who is Director of Integrated Energy Research at Wood Mackenzie.

"As launch timelines compress and costs decline, we are seeing the sector move from pilot projects to commercial scale operations across multiple markets simultaneously."

Companies including Tesla, Waymo, Baidu and Xpeng all adopted VLA technology in late 2025 and are scaling its application.

This technology allows AEVs to substitute costly rotational LiDAR sensors with lower-cost alternatives and high-dynamic-range cameras.

Waymo's expansion and Chinese providers

Waymo is leading American expansion with plans to operate in 27 cities by the end of 2026, up from just five.

This rollout is supported by a new lower-cost vehicle platform and strategic partnerships with Uber and Avis Budget Group.

China is the only market comparable to the US in scale.

Apollo Go, Pony.ai and WeRide have announced expansion plans across Beijing, Shanghai, Guangzhou, Wuhan and Shenzhen.

According to Wood Mackenzie, China's 15th Five-Year Plan, expected in 2026, will likely introduce new targets for the AEV supply chain with focus on Level 4 autonomy standards and data centre infrastructure.

Wood Mackenzie expects China to begin exporting its AEV technology globally in 2026, mirroring its position in solar panels and battery manufacturing.

Chinese firms offer cost-effective technology that is proving attractive to Transport-as-a-Service providers in Europe seeking to launch pilot programmes without significant capital investment.

Growing pilot programmes

London's complex road network will serve as a testing ground for European AEV adoption in 2026.

Waymo, Baidu and Wayve each plan to launch pilot programmes in the city during the year.

Transport-as-a-Service providers in the Netherlands, Germany and France are increasingly turning to Chinese technology for their pilot programmes due to cost-effectiveness.

The UAE is also emerging as a participant in the space, with approximately 130 autonomous vehicles currently being tested across the Emirates.

Baidu and WeRide are targeting fleet expansions to 1,000 vehicles by year-end in the country, aligning with the UAE's Vision 2030 goal of making 25% of all transportation autonomous.

The UAE's interest reflects broader economic diversification efforts and favourable regulatory environments that could be accelerating deployment timelines.

According to Wood Mackenzie, global AEV sales could expand tenfold between 2026 and 2030.

In the United States alone, the Transport-as-a-Service fleet will add 100,000 vehicles during this period.

Charging infrastructure will require substantial investment to support commercial operations, as autonomous vehicles in service require high-speed charging to minimise downtime.

Data centre expansion is another infrastructure challenge, as autonomous systems require continuous training using data from millions of miles of driving.

Power demand from AEV operations remains modest, but Wood Mackenzie's analysis suggests the sector could have significantly larger impact on commodity markets under high-case scenarios.