Buried in the reporting on Waymo’s lobbying surge is a line that stuck with me: critics have accused Uber of seeking regulatory advantage rather than competing on the product itself. Paraphrased, sure, but that accusation is the whole story in miniature. And my honest reaction, as someone who spends most of his week poking at AI tools to see which ones actually hold up, is that the critics are describing the industry norm, not an outlier.
Here is what we know. Waymo has doubled its lobbying spending as it goes head to head with Uber in the robotaxi market, with its efforts aimed at shaping regulatory policy. Separately, Streetsblog reported that Uber and Waymo combined have spent more than $15 million this year lobbying New York politicians. Ars Technica, the FT, and a lively Hacker News thread all picked it up. That is the factual floor, and it is enough to work with.
Why a Tool Reviewer Cares About Lobbying Filings
I review toolkits. I benchmark latency, I test failure modes, I try to break onboarding flows, and I write down what breaks. None of that captures the variable that decides which robotaxi you eventually ride in.
Autonomous driving is the rare AI category where the product cannot ship without permission. A coding assistant can be mediocre and still find users. A robotaxi that is not legally allowed to pick up a passenger in a given city has a market size of zero, no matter how good the perception stack is. That makes regulatory access a hard dependency, and once something becomes a hard dependency, companies spend on it like infrastructure.
So when Waymo doubles its lobbying budget, I do not read it as a scandal. I read it as a product decision. Someone looked at the roadmap and concluded that the marginal dollar buys more expansion in a statehouse than in a lab. That is a rational call. It is also a signal about where the real bottleneck sits, and that signal is more informative than any demo video.
What This Changes About How I Evaluate the Space
If you follow AI tools for a living, the lesson generalizes. A few things I have started weighing more heavily:
- Permission risk is product risk. Any tool whose core function depends on a license, a waiver, or an approval carries a failure mode you cannot test in a sandbox.
- Policy spend is a maturity indicator. Companies do not spend eight figures lobbying over a science project. The money says both firms believe the market is close enough to fight over.
- Incumbency can be manufactured. Rules written today decide who gets to operate in 2030. The best technical stack does not automatically win the rulemaking.
- Local matters more than national. The New York number is a reminder that this is being fought city by city and state by state, not in one sweeping federal moment.
Two Different Bets
The competitive framing here is interesting because Uber and Waymo are not the same kind of company making the same argument. Uber runs a network and needs supply flowing through it. Waymo builds and operates the vehicles. Both want favorable rules, but favorable means something different depending on which side of that split you are on. When two players with different structural interests are both spending heavily on the same regulators, the resulting rules tend to reflect whoever showed up more often, not whoever built the safer system.
That is the part I find uncomfortable. I would rather live in a version of this where the deciding factor is a public safety record you can audit. Instead the deciding factor increasingly looks like access, and access is purchasable in a way that engineering quality is not.
What I Would Watch Next
I am not going to pretend to know how this resolves, and the reporting does not tell us. What it does tell us is where to look. Follow the disclosure filings, not the launch events. Watch which cities open up next and check whether the timing lines up with a lobbying push. When a company expands into a market suspiciously fast, ask what happened in that market’s legislature six months earlier.
For readers here, the practical takeaway is small but real. When you evaluate an AI product that touches the physical world or a regulated industry, add a line to your checklist for how the company handles policy. Not because it makes the tool better, but because it determines whether the tool exists in your city at all. Right now that line item is running Waymo and Uber more than $15 million in one state alone. Worth knowing what you are actually looking at.
🕒 Published: