The announcement landed on Crypto Briefing early Tuesday: Uber had launched autonomous vehicle services in Zagreb, Croatia. No technical specifications. No partner name. No vehicle count. No safety driver disclosure. The most informative metric in that press release is the absence of data.
I have audited smart contracts for 400 hours in 2018. I modeled Compound Finance’s yield decay in 2020. I mapped Terra’s reserve flows in 2022. Every time a project hides operational details, it is because the details reveal fragility. Uber’s Zagreb launch is no different.
Context: The Platform’s Second Act
Uber sold its autonomous driving division (ATG) to Aurora in 2020. Since then, its strategy has been clear: become the platform that integrates third-party autonomous technology, not the developer. This is a capital-light model. It avoids the billions in R&D that Waymo and Cruise have burned. But it also means Uber’s autonomous future depends entirely on the quality and reliability of its partners.
In 2024, Uber partnered with Waymo for select cities in the U.S. and with Motional for Las Vegas. Europe, however, is a different regulatory landscape. The EU’s AI Act classifies autonomous driving as high-risk. Data must comply with GDPR. Local road conditions, signage, and driving culture vary massively from Croatia to Germany to the UK. A single pilot in Zagreb cannot solve for that complexity.
Core: The On-Chain Evidence… Of Absence
Let me treat this announcement like a smart contract. I will parse the fields that are present and the fields that are missing.
Present fields: “Uber,” “autonomous,” “Zagreb,” “Croatia.”
Missing fields: partner identity, vehicle model, sensor configuration, operational domain, safety driver protocol, pricing model, trial duration, user cap.
In crypto, when a project launches with a high TVL but no audit report, the market discounts it. The same logic applies here. Uber’s pilot is a closed-box transaction. I cannot verify the technology stack. I cannot stress-test the safety assumptions. Trust is a variable, not a constant. This launch asks for trust without providing a balance sheet.
Based on my industry knowledge, the most likely partner is Wayve, the UK-based autonomous driving startup that raised $1.05 billion in 2024 and counts Uber as an investor. Wayve uses an end-to-end learning approach, which is less reliant on HD maps and rule-based systems. This could be an advantage in Europe, where mapping every road is expensive. But end-to-end learning is also harder to audit. The safety case is a black box. That is a structural risk.
Zagreb was chosen for a reason. It is not Berlin, London, or Paris. It is a mid-sized capital with relatively simple traffic patterns, low regulatory overhead, and a government eager to attract tech investment. The cost of failure is low. The cost of learning is also low. This is a probe, not a beachhead.
Contrarian: Correlation Does Not Equal Causation
Some analysts will interpret this as a signal that Uber is accelerating its autonomous timeline. They will point to the stock price bump. I do not buy that.
Volatility is the price of permissionless entry. The stock price reaction to a small pilot is noise. The real signal is what Uber does next. If they announce a second city within six months – say, Munich or Stockholm – then we have a trend. If they reveal the partner and the safety data, then we have a verifiable claim. Until then, this is a press release, not a product.
The contrarian angle is that this pilot is not about technology or revenue. It is about regulatory positioning. Uber wants to show European regulators that it can operate safely under local laws. It wants to preempt Waymo, which has yet to enter Europe. By being first, Uber can shape the narrative and the rules. Yields attract capital; sustainability retains it. The yield here is regulatory goodwill. The sustainability will depend on how well the pilot performs.
Takeaway: The Next Week’s Signal
I will be watching for three data points. First, the partner name. If it is Wayve, the pilot is a test of end-to-end learning in a live city. If it is a lesser-known firm, the risk of technical failure increases. Second, the safety driver disclosure. If no safety driver is present, the pilot is operating at Level 4, which is a significant milestone. If a safety driver is present, this is Level 2+ and the commercial value is minimal. Third, the user feedback. Uber will collect trip ratings. If the average rating is below 4.5, the technology is not ready for prime time.
I have seen this pattern before. In 2020, DeFi projects launched with high APY and no vesting schedule. The yields attracted capital, but the sustainability was zero. Uber’s Zagreb pilot is a yield event. The question is whether the underlying protocol can sustain trust. The exit liquidity is someone else’s entry error. Do not be the liquidity.
The data vacuum is the most honest part of this announcement. It tells me what Uber does not want to say: that this is an early, fragile, and unverified experiment. I will wait for the audit results.