In Q2 2026, an AI shopping agent contacted 100 dealerships about the same vehicle. Within 24 hours, 92 responded. 33 provided full out-the-door pricing. The agent produced a ranked shortlist in minutes, with zero fatigue and zero social pressure. A human shopper doing the same work faces days or weeks of calls, follow-ups, and showroom hours.
Sit with the scale of that for a second. One buyer just ran a 100-store competitive bid process before breakfast.
The numbers
| AI Shopping Agent | Human Shopper | |
|---|---|---|
| Dealers contacted | 100 | 1 at a time |
| Responded within 24 hours | 92 | Wait, follow up, repeat |
| Full out-the-door price obtained | 33 | Sit down, negotiate, leave |
| Time to ranked shortlist | Minutes | Days or weeks |
| Fatigue / social pressure | Zero | Hours in the showroom |
That is not a faster consumer
The instinct is to read this as the next step in an old story. Customers got the internet, then they got TrueCar, now they have agents. More informed shoppers, same game.
Wrong read. That is not a faster consumer. That is a different kind of counterparty.
Every defense the traditional sales process relies on is calibrated for humans. The negotiation is built on stamina: the customer gets tired, hungry, emotionally invested in driving home today. The follow-up cadence is built on attention: people lose track, get busy, settle. The pricing strategy is built on opacity: most shoppers will not extract 33 comparable out-the-door numbers, because doing so costs a week of their life.
An agent has no stamina to wear down, no attention to lose, and no cost to asking 100 stores the same precise question simultaneously. It does not get embarrassed asking for the OTD breakdown three times. It does not fall in love with the blue one. It ranks, and its owner picks from the ranking.
The information advantage is finished
For most of automotive retail's history, the dealer held every number and the customer held a brochure. That asymmetry was not incidental to the business model. It was the business model.
The internet eroded it. Invoice prices went online in the late nineties, dealers fought it, and the direction didn't care. AI completes the job. When any buyer can deploy an agent that compares 80 or 100 stores in real time, the information gap that funded a century of margin is functionally closed.
The question that matters now: if information asymmetry in your market were eliminated tomorrow, what is left that you would still get paid for?
There are real answers to that question. Relationship. Experience. Local trust. Speed and ease of transaction. Service capacity an agent cannot replicate. But every one of them must be built deliberately. None of them can be assumed.
What the 33 should haunt you about
Inside the study, the most important number is not 92. It is 33.
Sixty-seven stores either refused to give an out-the-door price or could not produce one through their existing process. In the old game, that was a defensible move: force the appointment, control the conversation in person. In the new game, those 67 stores did not protect their margin. They removed themselves from the shortlist. The agent did not get frustrated or come in anyway. It ranked the 33 who answered and discarded the rest.
When the counterparty is an agent, opacity is no longer leverage. It is self-deselection.
How dealers should respond
The wrong response is trying to out-stall the agents. They do not tire. The right response is becoming the store agents rank first, while building the intelligence the agent cannot see.
Be machine-legible. Clean, complete, fast answers to pricing inquiries. Structured data on your site. If an agent asks 100 stores and 33 answer, be in the 33, and be the cleanest answer among them.
Match intelligence with intelligence. The shopper's agent knows the market. Your side needs to know the customer: full history, equity position, service record, the right offer for this specific person today. That is AI inward, built on data you own. The dealership that knows the most about the market loses to the dealership that knows the most about the customer.
Move the margin to defensible ground. If the gross depends on information the customer no longer needs you for, the gross is already gone and the P&L just hasn't caught up. Audit which products and fees are price-comparable on a phone in 60 seconds. Rebuild around relationship, experience, and service before the agents finish the job.
The era when exhausting the customer was a strategy is over. The dealership that knows the most, wins.
FAQ
What is an AI shopping agent? Software that shops on a buyer's behalf: contacting dealerships, requesting out-the-door pricing, comparing offers, and producing a ranked shortlist without human effort or fatigue.
What happened when an AI agent contacted 100 dealerships? In a Q2 2026 test, 92 of 100 dealerships responded within 24 hours, 33 provided full out-the-door pricing, and the agent built a ranked shortlist in minutes.
Should dealerships refuse to give pricing to AI agents? Refusing removes the store from consideration entirely. Agents don't get worn down or come in anyway; they rank the dealers who answered and discard the rest.
How should dealerships prepare for AI shopping agents? Become machine-legible with fast, clean pricing responses. Build owned customer intelligence the agent cannot see. Move margin away from products that are price-comparable in seconds.
Will AI agents kill dealership margins? They eliminate margin built on information asymmetry. Margin built on relationship, experience, service capacity, and customer intelligence remains defensible.
Todd Smith is the Founder and CEO of QoreAI and the author of The Intelligent Dealership: How AI and Data Transform Automotive Retail.