The Artificial Investor - Issue 34: The acceleration of robotaxis

My name is Aris Xenofontos and I am an investor at Seaya Ventures . This is the weekly version of the Artificial Investor that covers the top AI developments of the last seven days.

This Week’s Story: The acceleration of robotaxis?

The most interesting story of last week was WeRide ’s IPO. The broader Tech IPO market remains fairly dry and most investors have been focused on large private market fundraises. And then, one day last week, we didn’t just get a Tech IPO, but also the first pure-play robotaxi company IPO!?

China’s WeRide debuted on the Nasdaq last week, raising 120 million dollars at a 4.5 billion dollar valuation (c.110x its run-rate revenues ). The IPO took place just a few days after Tesla showcased its own robotaxi, the Cybercab , as part of its WeRobot event, and Waymo, the American market leader, announced a 5.6 billion dollar fundraise led by its parent, Alphabet.

Is the robo taxi trend accelerating? How far are we from mass adoption? How ready and safe is the technology???

?? We covered the basics of self-driving cars back in May 2024 in our 19th issue . There, we went through the Level 1-5 framework used to define driving automation, the capabilities of the technology, and an overview of the competitive landscape across the value chain. So, we won’t go through all of this now. But. we thought it was time for an update, given the first IPO of a robotaxi company ever.?

?? Early days

Autonomous rides are becoming a more common occurrence, particularly in certain parts of the world, such as the US West Coast and China’s largest provinces. Waymo, a subsidiary of Alphabet, is currently providing over 100,000 paid robotaxi rides per week in the US , double the volume we reported back in May 2024, implying an estimated c.100 million dollar revenue run-rate. To date, the company’s customers have covered nearly 15 million miles in driverless cars in San Francisco, Phoenix and Los Angeles. On the other side of the global map, Baidu’s Apollo Go had operated about 7 million rides up to August 2024, ending the second quarter of 2024 at a rate of 70,000 rides per week across Wuhan, Beijing, Shanghai, Guangzhou and Shenzhen, implying an estimated c.4 million dollar revenue run-rate (rides in China are 20 times cheaper than in the US West Coast).?

The top 7 players have a collective fleet of about 4,000 vehicles: AutoX leads with 1,000, followed closely by Waymo (700), WeRide (600) and GM’s Cruise (600); Baidu’s Apollo Go (400) and Pony AI (440) come next. The combined revenues of the top four players (Waymo, AutoX, WeRide and Apollo Go) are estimated at c.150-200 million dollars, which is 0.8% of Uber’s ridehailing run-rate revenues of 24 billion dollars .

?? The skies are open

Despite the early commercial stages of the market, investor appetite for the sector remains high. Just last week, Waymo secured 5.6 billion dollars in a new funding round by Alphabet (making Waymo its largest investment to date), a16z, Fidelity and T. Rowe Price, at an estimated valuation of 50 billion dollars (500x run-rate revenues). This brings Waymo's total capital raised to 11.1 billion dollars. Waymo is still the only company operating commercial robotaxi services in multiple major metropolitan areas in the United States.?

At the same time, WeRide raised 440 million dollars, 120 million through a Nasdaq IPO and the remaining through a private placement. The company has conducted autonomous driving tests and operations in over 30 cities across 7 countries worldwide for more than 1,700 days. WeRide remains the only robotaxi company licensed to operate in the UAE.

?? Checkmate

The strategic war in the sector is well underway. Having a look at the recent moves of the key players, it appears that different business models have emerged depending on: i) who manufactures the car, ii) who operates the taxi service and iii) who develops the autonomous-driving technology. Most players are diversifying geographically, given that the national markets are still small and limited by regulatory restrictions (more on that later). The car manufacturers and ridehailing companies have naturally gotten early in the game with both strategic investments and partnerships.

???? Regulatory update

Not much development on the American regulatory front, as there is still no federal law and citizens rely on state legislation, with California, Texas, Arizona and Georgia being the pioneers. It is expected that the Federal Motor Carrier Safety Administration (FMCSA) will publish a proposed law regarding commercial autonomous vehicles in December 2024.?

The EU is behind the US in terms of regulatory readiness for robotaxis at scale. There is EU law (EU 2022/1426) that promotes and regulates Level 3 automation (i.e. automated vehicles replacing occasionally the driver) on motorways, and allows for approval of fully-automated vehicles produced in small volumes (limited to 1,500 units annually). The EU is working towards enabling level 4 autonomy by 2025 under its "Europe on the Move" initiative. The most advanced country from a permit perspective is Germany, which has given licence for pilots to Mobileye ’s technology (retrofitted in German OEMs, such as VW) and Rimac ’s robotaxi service (in 2026) in cities, such as Munich.?

In May 2024, the UK introduced the Automated Vehicles Act 2024 (AVA 2024 ), which covers vehicles with Level 3 and Level 4 automation and obliges them to pass a "self-driving test" before being authorised for use on public roads. The UK government aims to have self-driving vehicles on the roads by 2026. Wayve and Oxa are trialling self-driving cars in London and Oxford respectively, and Rimac has obtained a licence to test its robotaxi service in Manchester (expected in 2026).?

?? Safety first

After a couple of setbacks in 2018 (Uber vehicle killed a pedestrian in Arizona ), late 2023 (Cruise vehicle killing a pedestrian in California ) and early 2024 (Waymo recalling 444 vehicles after two minor collisions in Arizona ), the industry has been boosted by positive safety studies that were released recently.?

A study by the University of Central Florida found that self-driving cars had lower accident rates than humans in work zones, traffic events and "pre-accident movements" like slowing down and proceeding straight. Autonomous vehicles were about half as likely to be involved in rear-end accidents compared to human drivers. Also, self-driving cars performed better in rain or fog, with about one-third the crash risk of human drivers. Overall, human ridehailing drivers had a crash rate of 50.5 crashes per million miles, while self-driving cars had 23 crashes per million miles. On the other hand, autonomous vehicles struggled more than human drivers at dawn or dusk (5 times more likely to be involved in an accident) and when making turns (nearly twice as likely to be involved in an accident).?

Probably not as objective, but still another data point. Waymo published data showing that its robotaxis were 3.65 times safer than human drivers in Phoenix and San Francisco, with 0.82 injury-reported crashes per million miles compared to 2.92 for human drivers.?

In an attempt to eliminate safety risks and broaden the scope of use cases, the technology has also been advancing recently. Waymo unveiled its 'generation 6' autonomous driving system that enhanced capabilities in various weather conditions, AutoX released its proprietary Gen5 fully-driverless system that is capable of handling complex traffic situations in urban centres, Baidu announced the release of a new version of its autonomous-driving platform (Apollo 10.0) that enables global road coverage, and WeRide developed "WeRide One" with advanced perception, lower latency and a full stack of software, hardware and infrastructure.

Nevertheless, for the time being, all robotaxi companies operate with a remote driver, who is ready to take control of the vehicle when the technology alerts them to do so.?

?? For a better world

Wrapping up, it feels like the robotaxi industry has accelerated since we last touched on the topic in May 2024. Further money has been pumped into the sector, we saw the first IPO of a pure-play robotaxi company, various US states and Middle East countries have been giving permissions to service providers to operate, and Elon Musk’s entry into the market has heated up competition.?

Public opinion also feels that it has been shifting, as the focus has been moving away from job losses and accident risks, to the positive economic and societal impact.?

The University of Texas estimates robotaxis’ potential economic impact in the US of 1.2 trillion dollars or 3,800 dollars per American per year. King's College London highlights the potential benefits for socially-excluded groups, such as people with disabilities and the elderly, which accounts for 22% of the UK's population. The University of Toronto analyses dozens of existing studies and explores the implications on vehicle utilisation (up, from 5% currently to potentially 75%), car ownership (reduction by 25%), land use (20% less parking space) and employment (millions of truck driver jobs lost). McKinsey estimates that autonomous driving could create 300 to 400 billion dollars in revenue by 2035, implying a nearly 100% compound annual growth rate. The Compact magazine estimates that for every 1% reduction in road safety incidents in the US, the country could save 8 billion dollars annually, and suggests that self-driven cars could free up to 50 minutes per day per passenger (that were previously dedicated to driving).?

?? We are very excited about the future of autonomous vehicles, as they have the potential to significantly impact various aspects of society and the economy, including transportation costs, urban planning, employment, safety and accessibility. Stay tuned for more updates on the sector.

??? Fun things to impress at the dinner table

Shocking. Not a fun thing at all, but certainly some food for thought. A 14-year old American committed a suicide and his mother sued Character.ai , as her son was spending many hours with the AI companion app prior to the tragic incident.?

See you next week for more AI insights.