

There's a gap opening up in dealership AI, and this week's news finally puts a number on it. Everyone bought something. Almost nobody can prove it worked. That's not a reason to panic — but it is a reason to stop buying and start checking.
Cox Automotive's new AI in Auto Retail Tracker is the first real survey data on dealer AI adoption, and it should sting a little. 82% of dealers say they're using AI somewhere in the store. Only 22% of the ones expecting a sales or revenue bump actually got one. That's not a rounding error — that's four out of five dealers who bought a tool, checked a box, and are still waiting for it to do anything measurable.
Meanwhile the ground is moving under them. 63% of shoppers say they plan to use AI on their next vehicle purchase. Only 29% of dealers have changed anything about how they show up when that AI goes looking for cars. So the adoption gap isn't really a technology problem — it's a "we bought a chatbot and called it done" problem. The tools dealers are buying are mostly aimed at the wrong layer: customer-facing widgets that make the website feel modern, while the plumbing that actually determines whether a shopper's AI assistant finds the car at all goes untouched.
For an operator, the lesson is blunt: adoption without measurement is just spend. If you can't tie an AI tool to a lead, a close, or a saved hour, you don't have a strategy, you have a subscription.
The next twelve months separate two kinds of dealers: the ones who bought AI to look current, and the ones who audited whether their data can actually be found by it. Cox's numbers suggest most stores are still in the first camp. The stores that check their feed, fix what's broken, and can point to an actual number next quarter will be the case studies everyone else quotes — skeptically, as always, until they see the receipt.

ACV Auctions and DriveCentric are linking service-drive inspections directly to trade-in sourcing and fixed-ops upsell, while Numa's acquisition of SocketTime aims to sync customer-facing AI with real-time shop capacity. Both moves point the same direction: stop treating service and sales as separate systems that don't talk to each other.

One deal dominates this week: Numa's acquisition of fixed-ops software company SocketTime, aimed at moving AI past the front counter and into repair orders, dispatch, and technician capacity. It's the clearest sign yet that voice AI vendors know "answering the phone" was never the hard part — running the shop behind it is.