TechCrunch Mobility: A new, bigger AV boom is here

Trucks VC received pitches from 81 new autonomous-vehicle startups so far in 2026, two-thirds more than last year and above the 2016–17 boom-year levels; the firm attributes the surge partly to AI foundation models lowering development barriers. Re:Build raised $2.1 billion, while Bedrock Robotics raised $600 million at a $4.1 billion valuation; DoorDash also agreed to acquire Instacart for $17 billion. Other updates included Rivian raising its 2026 delivery floor from 65,000 to 67,000 vehicles and GM agreeing to a $220 million settlement with 36 states over allegations concerning driver data.

Analysis

The investable signal is not the startup count; it is a potential shift in AV economics. Open models can lower prototype costs, but they do not remove the harder bottlenecks: safety validation, permits, insurance, fleet utilization and reliable operations. That weakens software-only differentiation while increasing the value of compute, testing infrastructure and partners able to manufacture and distribute at scale. NVIDIA (NVDA) could benefit from greater inference demand, but open-source adoption alone does not guarantee its hardware captures that spend.

Waymo’s manufacturing path with Toyota, plus the possibility of licensing, would broaden Alphabet’s (GOOG) routes to monetization beyond operating its own fleets. That is strategically positive over 6–18 months, but licensing remains prospective; production and deployment milestones matter more than partnership headlines. Kodiak AI (KDK) has a more concrete but distant catalyst: the stated 2028 delivery start leaves substantial execution and commercialization risk, so the agreement supports validation, not near-term revenue certainty.

The reported DoorDash–Instacart transaction needs confirmation of terms and regulatory path before treating it as an arbitrage or earnings catalyst. If completed, denser grocery demand could improve marketplace economics, but integration, deal financing and competition review could offset synergies. Separately, the small number of AI-agent orders cited by DoorDash argues against assigning material near-term value to agent-led commerce.

Contrarian view: lower startup costs may create more pitches, not more durable winners. Over the next 1–3 months, watch binding deployment commitments, safety/regulatory milestones and disclosed transaction terms—not funding valuations. Failure to convert pilots into repeatable, permitted operations would falsify the bullish AV scaling thesis.

Key Decisions for Investors

• Avoid a broad AV-startup/AI basket trade: cheaper model access likely expands the field faster than it improves the economics of scaled deployment.
• Treat KDK as a high-risk, long-dated watch or small speculative position, not a near-term earnings trade. Reassess on evidence of regulatory clearance, deployment milestones and repeat orders; delay or cancellation of the 2028 program would falsify the thesis.
• Maintain a constructive 6–18 month view on GOOG’s AV optionality, with Toyota’s manufacturing role potentially widening distribution. Do not capitalize prospective personal-vehicle licensing until terms and deployment plans are disclosed.
• Put DASH/CART deal exposure on a verification alert rather than trading headline terms: confirm the agreement, consideration, financing and regulatory timetable first. If confirmed, compare achievable order-density and cost benefits against integration and antitrust risks.