Cannonball Run record holder Alex Roy on EVs, autonomous vehicles, and investing in deep tech

Summary

Alex Roy, Cannonball Run record holder, former Argo AI executive, and managing partner at New Industry Venture Capital, offers a practitioner's lens on EVs, autonomous vehicles, and the competitive threat from China that most legacy auto executives are only beginning to take seriously.

EV Charging Infrastructure: Still a Structural Problem

The non-Tesla charging network remains a significant drag on broader EV adoption. Handshaking failures, payment friction, and inconsistent uptime are systemic issues that no American charging provider outside Tesla has fully resolved. Roy frames this as a 100-year-old problem repeated: when internal combustion vehicles first scaled, Coca-Cola underwrote the buildout of regional gas station networks. Today's EV charging players have largely ignored that historical playbook.

The recent Cannonball EV record, set by a second-generation Porsche Taycan, illustrates the nuance. The Taycan holds a shorter EPA range than a Tesla Model S Long Range, but the Electrify America network's recent infrastructure upgrades enabled charging speeds that surpass Tesla's supercharger rates. The Taycan's battery chemistry sustains high charge rates up to approximately 50% state of charge, versus Tesla's cliff at 20%. That charging architecture, not raw range, is what broke the record. Roy expects the next-generation Tesla or a Lucid to reclaim it.

The EV Tax Credit Expiration Is a Minor Variable in a Larger Problem

The EV tax credit is set to expire in September, which will slow adoption and push prices higher for American EV buyers. Roy treats it as a second-order concern relative to the structural competitiveness gap with China. Chinese manufacturers have integrated vehicle hardware, software, and charging infrastructure into a coherent system that no American OEM outside Tesla has matched. Jim Farley at Ford is credited with genuine curiosity, having personally debriefed Roy over dinner on the Tesla Model 3 after Roy's last electric Cannonball record. The Mustang Mach-E reflects some of those learnings. But the honest assessment is that outside Tesla, Lucid, and Rivian, no American EV product is currently exportable at competitive price and quality into third-country markets against a Chinese alternative.

The parallel to the 1970s Japanese auto invasion is explicit. The dismissal then was that foreign manufacturers couldn't truly innovate. The result was a structural reshaping of the American auto industry. Journalist Kevin Williamson, who has driven current Chinese EVs extensively, characterizes them as superior in build quality to anything sold in the US, including Teslas.

Autonomous Vehicles: Waymo and Tesla Are Not in a Zero-Sum Race

The framing of Waymo versus Tesla as a winner-take-all contest misreads the market structure. Roy draws the analogy to the elevator industry: no one can name an elevator company other than Otis, yet the big five elevator companies are all large profitable businesses. Three to five AV companies will coexist.

The actual strategic tension is about business model timing. Waymo needs to move into hardware licensing and miniaturize its sensor stack to become invisible inside third-party passenger vehicles before Tesla can scale robo-taxi operations at a commercially meaningful level. Tesla needs to deploy a system that is perceptibly safe to the public and build robo-taxi density before Waymo can encroach on personal vehicle ownership. Neither is there yet, and both face execution risk specific to their respective models.

The Adjacent Opportunity Most Investors Are Missing

The job displacement narrative around autonomous vehicles is, on Roy's read, historically illiterate. Elevator operators went from 100% penetration in 1946 to 1% by approximately 1970, yet total employment across the elevator industry is larger today than at peak operator headcount. The same dynamic applies to transportation.

The more actionable investment insight is that an entire category of startups providing services inside autonomous vehicles does not yet meaningfully exist. The specific example he surfaces: elder care companionship layered onto Waymo or Uber infrastructure, connecting older or mobility-limited passengers with trained human attendants for errands. Most current Uber drivers, he argues, would be better deployed in that role than behind a wheel. New Industry Venture Capital focuses on deep tech, but Roy flags manufacturing and energy as the highest-concentration opportunity areas within the EV supply chain, noting that the marquee bets on OEMs themselves are largely taken and expensive.

The Policy Gap

Roy cites Ela Dinski of the Foundation for American Democracy on the need for what she calls a Pentagon for industrial policy, not a command economy, but a more coordinated national framework for identifying and scaling export-competitive technologies. The current dynamic, Tesla and Waymo consuming the domestic policy conversation while China builds export market share, is the wrong frame. The US needs national champions exporting at scale, and the internal political debate over EV incentives is a distraction from that longer-horizon competition.