Beginners in AI

The Mobility and Transportation Frontier

James Swierczewski's avatar

James Swierczewski

·

AI Summary: Waymo runs roughly 250,000 paid autonomous trips per week in Phoenix, San Francisco, Los Angeles, and Austin in 2026, with no human behind the wheel. Mercedes-Benz sells the only legally Level-3 production car in the United States. EV battery factories worth $200 billion are under construction across North America. Joby and Archer are flying Type-Certificate eVTOL prototypes. This hub is the beginner’s guide to mobility technology in 2026: where the future of moving people and goods actually is, separated from the marketing.

If you only know one number about modern mobility, make it this one: 250,000 paid autonomous rides per week. That is roughly Waymo’s pace in mid-2026 across four US cities, with no human behind the wheel. Five years ago the same number was effectively zero. The cars that drove themselves through downtown Phoenix in 2020 had a safety driver in the front seat. Today they do not. The cars are doing the work.

That is one of dozens of mobility numbers worth knowing in 2026. EV battery factories worth $200 billion are under construction across North America. China sells more EVs than the rest of the world combined and ships them at prices Western automakers can’t match. Mercedes-Benz quietly became the first manufacturer to sell a Level-3 production car in the US (eyes-off, hands-off, on certain highways, in good weather). Joby Aviation and Archer Aviation have FAA type-certificate paths for their electric vertical takeoff vehicles. Aurora launched commercial driverless trucking in Texas in 2024. Almost everything about how people and goods move is getting rebuilt, mostly out of sight of the average reader.

This hub is the Beginners in AI guide to mobility in 2026: real status, honest timelines, the AI underneath, and the human choices that decide what happens next.

Where to start

Special Report · Anchor Reading

The Real State of Self-Driving in 2026

Our flagship Special Report: where Waymo, Tesla, Aurora, and the Chinese L4 players actually are, what counts as Level 2/3/4/5, and what the next five years will look like. 18-minute read, eight charts, free PDF download.

Read the Special Report →

What this hub covers

  • Self-driving in 2026. Waymo’s commercial scale-up. Tesla FSD reality vs the marketing. Mercedes Level 3 in the US. Cruise’s collapse and what it taught the industry. Chinese L4 (Pony.ai, WeRide, Baidu Apollo). The deep dive lives in our Special Report.
  • EV battery wars. BYD’s pricing dominance. CATL’s scale advantage. Tesla’s 4680 cells. The Korean trio (LG, Samsung SDI, SK On). Solid-state timelines (Quantumscape, Samsung SDI, Toyota). Sodium-ion as the cheap-everywhere alternative.
  • The EV charging buildout. Tesla Supercharger network opening to other automakers. NACS standard adoption. EV Charge Network slow-pace reality. Why charging is now the actual bottleneck, not range.
  • eVTOL (the flying-car thing) in 2026. Joby Aviation, Archer Aviation, Lilium (in bankruptcy), Vertical Aerospace, EHang. FAA Type-Certificate timelines. Why noise is the real problem. The realistic urban-air-mobility timeline.
  • Autonomous trucking. Aurora’s Texas commercial routes. Kodiak. Plus AI. Why long-haul highway driving was the right place to start automation.
  • Hyperloop status. Honest assessment of where the hyperloop bet sits in 2026. Spoiler: not great. Why high-speed rail is the more interesting story.
  • Autonomous shipping. Mayflower, Yara Birkeland, the Chinese pilots. Cargo ships running with skeleton crews and AI-driven navigation.
  • Public transit and freight rail. The unsexy infrastructure that moves most of the world. Why electrification of freight rail is one of the most leveraged decarbonization moves available.

The mobility story in five numbers

  • 250,000. Approximate weekly paid Waymo rides in mid-2026. From basically zero in 2020.
  • $200 billion. Approximate value of EV battery factories under construction in North America since the Inflation Reduction Act. Most are in red states.
  • $10,000. The price gap between a competitive Chinese EV and a competitive American EV in the same class. A real strategic problem for the US auto industry.
  • 1 in 5. The share of new car sales globally that are electric (battery EV plus plug-in hybrid) in 2025. From under 1 in 100 a decade ago.
  • ~80 decibels. The noise produced by current eVTOL prototypes at takeoff. About like a garbage truck. The unsolved problem keeping flying cars out of cities.

How AI fits in

  • Self-driving is, at its core, an AI problem. Modern autonomous-vehicle stacks are end-to-end neural networks trained on billions of miles of driving data. Waymo, Tesla, Aurora, and the Chinese players all use different architectures, but the underlying technology is one field.
  • Battery materials are AI-discovered. Google DeepMind’s GNoME system found 2.2 million new stable crystal structures, many of them candidates for the next generation of batteries. AI has compressed decades of materials chemistry into months.
  • Logistics routing. Modern freight, last-mile delivery, and ridehail dispatch all run on machine-learning optimization. The reason your food shows up faster than it used to is mostly software.
  • Vehicle design. Generative AI helps automotive designers iterate exterior and interior concepts in hours that used to take weeks. The downstream engineering simulations (crash, aero, NVH) all run on AI-accelerated solvers now.
  • Predictive maintenance. The data coming off a connected fleet (your Ford, your Tesla, a UPS truck) feeds models that flag failures before they happen. Maintenance is becoming a real-time service.

The Beginners in AI position on mobility

Roughly 40,000 Americans die in car crashes every year. The combination of cleaner cars, smarter roads, autonomous vehicles that are already safer than the median human driver in conditions they handle, and electrified freight is one of the largest improvements to daily life that technology can deliver. We are going to write about every step of it with the enthusiasm it deserves.

And there is a tension we will keep writing about honestly. Mobility technology touches the choices that define daily life: where you live, what you can afford, whether you can work where you want to work, whether your kid can get to school safely, whether your parents can keep their independence as they age. The technology should serve those choices, not override them. We will write honestly about which mobility shifts are improvements and which are tradeoffs the industry would rather you not notice.

Two specific things to remember. First, the cheap car you actually need probably still exists, and if the auto industry quietly gets rid of $20,000 new cars to chase software-locked $60,000 ones, that is a problem. Second, public transit, walkability, and bike infrastructure remain the cheapest and most pro-human forms of mobility ever invented. We will write about all of them.

Frequently asked questions

Should I buy an EV right now?

Yes for most situations, no for some. EVs are an excellent choice if you have home charging, drive mostly in town and on highway, and keep cars for many years. EVs are a worse choice if you live in a multi-family building without dedicated charging, drive long road-trip distances regularly, or live somewhere with sub-zero winter temperatures (range hits 20-40 percent). We will write a dedicated buying guide.

When can I buy a self-driving car?

You can ride in one this week if you live in Phoenix, San Francisco, Los Angeles, or Austin (Waymo). You can own a partially-autonomous one today (Mercedes Drive Pilot for Level 3 on certain highways, Tesla Autopilot/FSD for Level 2 with strong supervision required). Fully autonomous (L4/L5) consumer ownership is years away from broad availability. We do not know which company gets there first.

Are eVTOLs (flying cars) ever going to be real?

Real in the sense of working aircraft: yes, Joby and Archer are flying. Real in the sense of widely available, affordable urban mobility for non-rich people: probably not this decade. The bottleneck is noise, regulation, and price per ride. We will write about each piece honestly.

Is Tesla’s Full Self-Driving really self-driving?

Currently no. Tesla FSD is a Level 2 advanced driver-assistance system. The driver is legally responsible at all times and the car requires hands-on/eyes-on supervision. Tesla has stated intentions to release a Level 4 robotaxi service and an unsupervised version of FSD for owners. Neither has shipped. The honest framing in 2026 is “FSD is genuinely impressive Level 2 software, with marketing that implies more than it delivers.” Our Special Report covers the details.

Where do I read first?

Start with our Special Report on self-driving. For EVs, InsideEVs and Electrek. For the auto industry, Automotive News. For aviation, FlightGlobal. For policy, Bloomberg Green’s EV coverage.

Sources and further reading

You might also like

Discover more from Beginners in AI

Subscribe now to keep reading and get access to the full archive.

Continue reading