Foundation models drive a new wave of autonomous vehicle startups
Trucks VC says it has received pitches from 81 new autonomous vehicle startups so far in 2026, a total that is two-thirds higher than last year and already exceeds the pace seen during the sector's 2016 and 2017 investment boom. The early-stage investor argues that a second, larger autonomous-vehicle wave is forming around advances in artificial intelligence.
Puneeth Meruva, a partner at Trucks VC, said foundation models are changing the technical path toward end-to-end autonomy systems. He described the shift as a move from AV 1.0 and AV 2.0 ambitions toward an architecture in which models can support the long-sought end-to-end approach.
Open models lower the cost of entry
Meruva said foundation models can generalize across different environments and scenarios, potentially allowing autonomous systems to be deployed more quickly and at lower cost. The arrival of open-source models such as Nvidia's Alpamayo and Qwen's general-purpose vision-language-action models has made it easier for new companies to begin building products.
Trucks VC is seeing startups reach deployment within a few months after raising less than $1 million, Meruva said. That is a markedly different starting point from earlier autonomous-driving efforts, which often required much larger teams, long development cycles and extensive capital before a service could be tested or launched.
Robotaxis return to the startup agenda
The investor expected many founders to concentrate on construction, warehouses, agriculture and mining, where automation can be limited to more defined operating environments. Those applications are still appearing in pitches, but Meruva said a growing number of teams are proposing pure-play robotaxi businesses.
Several of those companies position themselves as competitors to Waymo. Meruva's view is that smaller autonomy providers may not need to challenge Waymo globally to build a business. They could instead serve particular regions or route types, a market structure that aligns with the contrasting scale-up approaches in contrasting autonomous vehicle scale-up approaches and leaves room for specialised operators.
Funding signals extend beyond software
The broader mobility market is also attracting substantial capital. Bedrock Robotics raised $600 million in a Series C round valuing the autonomous-vehicle technology company at $4.1 billion, while Kodiak AI agreed to deliver 75 self-driving trucks to DTL Transport from 2028 in its first binding commercial deal for driverless trucks on public highways.
Waymo and Toyota have also formalised a manufacturing partnership. Toyota will build AV-ready bZ4X vehicles and integrate Waymo's self-driving system at its factory in Japan for robotaxi services in Japan, the UK and other international markets. The companies indicated the arrangement could eventually support a licensing model for personally owned vehicles.
What businesses should measure
A larger field of startups does not by itself establish commercial maturity. Businesses evaluating autonomous-driving suppliers should focus on the declared operating domain, the vehicle and manufacturing plan, evidence of deployment, and the capital required to maintain operations. The current wave makes those checks more important because lower-cost model development is bringing more contenders into the market.