WeRide: The Cheapest Pure-Play Bet on Autonomous Driving

Recent picks: XPeng (XPEV) | UltraGreen (ULG.SI)

Open your Uber app in downtown Abu Dhabi right now. Hit “Autonomous.” A driverless car will show up — no one in the front seat. You get in, say your destination, and it takes you there. This isn’t a demo. It’s a live commercial service you can pay for today.

The company behind it trades at US$7.09. I own shares. I’ve been adding throughout February. Here’s why.

What WeRide Actually Does

WeRide (NASDAQ: WRD, HKEX: 0800) makes Level 4 autonomous driving technology. Level 4 means fully driverless — no safety driver, no one ready to grab the wheel. The car handles everything. They’ve been running this commercially in Guangzhou, Beijing, and Shenzhen since 2019, across some of the most chaotic traffic in the world. Over 40 million autonomous kilometres logged.

Most people only know WeRide for robotaxis. But the same AI platform powers four other products — autonomous buses, delivery vehicles, and robotic street sweepers. That last one might sound boring, but if you walk along Marina Coastal Drive, the Esplanade, or Jurong Lake Gardens, you’ve probably already seen their cleaners working in the early hours. WeRide has been running those commercially in Singapore since late 2024. Most investors don’t realise this business exists. Every kilometre the sweeper drives makes the robotaxi smarter — it’s all one brain.

Why 2025 Matters

WeRide IPO’d on Nasdaq in October 2024 at US$15.50 — the world’s first publicly listed robotaxi company. The stock shot to nearly US$44 after Nvidia disclosed a stake, then crashed back down as the hype faded. It dual-listed on HKEX in November 2025, so you can buy it on either exchange now.

The fall made sense — typical post-IPO hype and selloff, compounded by investors rotating into the Mag 7 as the AI trade concentrated upstream. The IPO price already reflected the loss-making years.

But something shifted in 2025. Q3 revenue jumped 144% year-on-year to US$24 million. Robotaxi revenue specifically — the core thesis — grew 761%. Gross margin went from 6.5% to 33% in one year. And they cut operating expenses by half while growing revenue. That combination doesn’t happen by accident. And the timing matters — the 2026 investor playbook is increasingly robotaxis and robotics, and WeRide is one of the few names where the fundamentals are actually catching up to the thesis.

They’re also sitting on US$760 million in cash. That’s not a trivial number for a company at this stage — it means they don’t need to raise money desperately anytime soon.

Next earnings are 13 March 2026. That’s the next thing I’m watching.

How They’re Expanding

WeRide’s playbook is to go asset-light globally — they bring the tech and the cars, local partners handle everything else. What I like is how they adapt the model depending on where they are.

In Abu Dhabi, they partnered with Uber and launched in December 2024. By 12 February they’d already expanded to cover roughly 70% of the city’s core. The fleet has quadrupled. Each car is breaking even at 12 trips a day — they’re averaging 20 to 25. By 2027, WeRide and Uber are targeting 1,200 robotaxis across Abu Dhabi, Dubai, and Riyadh. Right now they have 200. If each car generates US$40,000 annually for WeRide, that’s US$48 million just from the Middle East.

Scaling that fast only makes sense if the unit economics hold. In July 2025, WeRide launched a new computing platform co-developed with Lenovo and powered by Nvidia’s latest chips — cutting vehicle hardware costs by 50%. The impact showed up directly in Q3 2025 margins. Lower cost per car means more cities, faster.

In Singapore, the partner is Grab, which is also putting money into WeRide as a strategic investor (deal expected to close H1 2026). The autonomous Ai.R shuttle is already testing in Punggol — Singapore’s first autonomous residential shuttle service — with public rides going live in weeks. Singapore is just the beachhead. Grab is investing in WeRide as part of a multi-year partnership to expand autonomous mobility across Southeast Asia — eight countries including Indonesia, Malaysia, Thailand, Vietnam, and the Philippines, where Grab is already the dominant superapp.

Europe is a different model entirely. WeRide isn’t going through ride-hailing platforms there — they’re winning public transit contracts.

In Belgium their autonomous bus runs on a regular De Lijn public bus route. In France a fully driverless shuttle connects a business park to a TGV train station. In Switzerland the Canton of Zurich and national railway SBB are paying clients — and WeRide got Switzerland’s first-ever driverless robotaxi permit in November 2025.

Getting an autonomous driving permit isn’t like filing a business registration. Abu Dhabi’s city-level commercial permit took WeRide four years of logged kilometres, safety testing, and regulatory trust-building to secure. Switzerland’s first-ever driverless robotaxi permit didn’t happen overnight either. Eight countries worth of permits is years of work that competitors can’t shortcut — and that’s before they’ve even turned a wheel. Waymo, for all its scale, is only operating commercially in the US — their London launch is still pending regulatory approval. No other company comes close to WeRide’s regulatory footprint.

What Worries Me

I’m not going to pretend this is a safe bet. Here’s what keeps me honest.

The losses are real. WeRide bled money every year from 2020 to 2024. Q3 2025 was encouraging — net loss narrowed 70% year-on-year — but one good quarter doesn’t fix years of cash burn. The July 2025 Lenovo/Nvidia hardware partnership already cut costs meaningfully, and it’s showing up in the margins. But full company profitability is still a 2028-2030 story — they need to scale to tens of thousands of vehicles before the numbers truly flip. The US$760 million buys time, not certainty.

Most of the revenue is still in China. Roughly 55-65% of the fleet is there. China risk is real — geopolitics, regulation changes, anything that hits the domestic business hits WeRide hard. The international story is the hedge, but it’s still early days.

They’re doing a lot at once. Eleven countries, three different partnership models, pre-profit. That’s a lot to manage. Permits get delayed. Governments change priorities. Partners have their own agendas. Fast expansion is a strength right up until it becomes a liability.

The competition isn’t sleeping. London alone tells you how fast things are moving — Waymo targeting it as its first move outside the US, Wayve raising US$1.2 billion to launch robotaxi trials there with Uber, Baidu’s Apollo Go partnering with both Uber and Lyft for UK trials. WeRide isn’t in London — they’re already operating commercially in Abu Dhabi, Singapore, and across Europe with robobuses. Pony.ai is pushing into Europe through Stellantis and Singapore through ComfortDelGro. But the message is clear: serious money is chasing international AV deployment fast. WeRide has the head start — in permits, in operating cities, in revenue. But these are not small competitors.

Should I Just Buy Alphabet or Baidu Instead?

I thought about this. Both Waymo (inside Alphabet) and Apollo Go (inside Baidu) are genuinely impressive operations. But buying Alphabet for Waymo is like buying Amazon because you believe in cloud computing, or Alibaba because you believe in fintech — the core business will always dominate how the stock moves.

I actually rode a Waymo in San Francisco — smooth, quiet, no jerky braking, genuinely impressive. But Alphabet is a US$3.7 trillion company. If Waymo doubles in value, Alphabet shares move maybe 3-4%. Every time Google Search misses — which has nothing to do with robotaxis — my “autonomous driving investment” takes the hit.

Baidu’s Apollo Go peaked at ~300,000+ fully driverless rides a week in Q4 2025 and already profitable per vehicle in Wuhan. But the core search business is in structural decline, and full-year profits plunged ~76% in 2025, weighed down by a one-off impairment charge. Apollo Go is a bright spot inside a shrinking company.

Waymo can’t be bought. Apollo Go is buried inside Baidu.

WeRide Key Financials

What I Think

I hold about 4.8% of my portfolio in WeRide — started at around 2.7% in January and have been adding through February. I mentioned the initial position in my January 2026 portfolio review.

At 24x EV/Revenue, WeRide looks expensive at first glance. The self-driving sector median is 2.1x — but that includes legacy car companies and ADAS suppliers who are nowhere near full autonomy. Among high-growth AI companies, 20-30x is normal. WeRide sits right in line.

The more honest comparison is against the other pure-play AV names. Waymo just got valued at 360x EV/Revenue in its latest funding round. Pony.ai trades at 63x. WeRide, the only one with commercial operations across three continents and permits in eight countries, trades at 24x. Both WeRide and Pony.ai are growing revenue at roughly 40% year-on-year. The valuation gap is hard to explain on fundamentals — which is why I think WeRide is mispriced.

For context: Wayve, still pre-commercial, was valued at US$8.6 billion in its latest funding round — with no revenue to speak of yet.

If you apply Pony.ai’s multiple to WeRide’s revenue, the implied enterprise value is around US$4.4 billion. WeRide’s current EV is US$1.7 billion. That’s not a price target — just a way to see how wide the gap is.

GlobalData named WeRide one of five companies the AV market will consolidate around globally — alongside Waymo, Apollo Go, Zoox, and Pony.ai. Morgan Stanley is projecting 103% annual growth in autonomous vehicle miles through 2032. In Q3 2025 alone, Guangzhou trips grew 4x and Beijing trips grew 8x quarter-on-quarter.

I added more in February because 2026 feels like the year things start moving — Abu Dhabi went fully driverless, downtown expansion just launched, and Grab is backing WeRide to take on Southeast Asia. WeRide is no longer a “wait and see” story.

This is a high-risk bet. Pre-profit, China-heavy, fast-expanding, competing against companies with far deeper pockets. I’m not trying to call the bottom — but I think the risk/reward at current prices is skewed to the upside over a 3-5 year horizon. If 2026 is the inflection year that the numbers are pointing to, then US$7.09 looks like the reset price — not the top.

This article represents my personal opinion and investment decisions. It is not financial advice or investment research. I hold positions in securities discussed. Always do your own research and consult a qualified financial advisor before making any investment decisions. WeRide trades on NASDAQ (WRD) and HKEX (0800.HK).