A new number came out last week. Reynolds & Reynolds, 57% of dealership staff now report using AI.
That's a remarkable figure. Eighteen months ago it would have been unimaginable. And it should make every operator in this industry stop and ask one question: using it to do what, exactly?
Because adoption is not implementation. Clicking a ChatGPT tab between service calls is not a strategy. Neither is a vendor contract that nobody trained your staff to use. The gap between "we have AI" and "AI is producing a measurable return" is where most dealerships are living right now — and where the real competitive leverage is for the ones willing to close it.
That's what this newsletter is about. Every issue: what's real, what's working, and what the numbers actually say.
— Dave Foy, Founder, Automotive AI Summit
IN THE INDUSTRY
The adoption gap is wider than the headline suggests
Reynolds & Reynolds data shows 57% of dealership staff self-report AI use — but the same research found that fewer than 1 in 5 could describe a specific workflow where AI changed their output. Usage and integration are two different things. The vendors know this. Most dealers don't yet.
F&I compliance gets an AI module
Reynolds quietly rolled out an AI-assisted compliance monitoring layer inside their F&I platform. It flags deviations from disclosure scripts in real time. The pitch is liability reduction. The actual value is consistency — particularly for high-volume stores where manager turnover creates compliance gaps. Worth watching.
One Southeast dealer cut inbound scheduling calls 41%
A multi-rooftop operator in the Southeast deployed an AI scheduling assistant across their service lanes. Inbound call volume dropped 41% in 90 days. Advisors now handle fewer transactional calls and more upsell conversations. The tool didn't replace anyone — it changed what the existing staff spent time on. That's the pattern worth replicating.
SIGNAL FROM OUTSIDE
What hospital systems figured out that dealers haven't
When hospital systems started deploying AI for patient scheduling and triage, the ones that succeeded did one thing differently: they named an owner before go-live. Not the vendor. Not IT. A specific person in operations whose performance was measured against the tool's outcomes.
The ones that failed treated AI as software infrastructure — someone else's problem once it was installed. The tool worked. Nobody was accountable for the results.
Dealerships are running the same playbook right now. The question isn't whether your AI tool works. It's who owns the outcome.
WHAT'S ACTUALLY WORKING
Declined service work follow-up
Dealers running automated AI follow-up on declined service work are seeing return rates of 15–18% within 30 days. That's not a marketing campaign — it's a service advisor workflow that most shops never had the bandwidth to run manually. The jobs were always there. The follow-through wasn't.
The setup: AI flags declined work at write-up, sends a personalized follow-up at day 7 and day 21, routes hot responses to the advisor. No additional headcount. The stores seeing the best results trained their advisors to expect the callback, not be surprised by it.
THE NUMBER
34%
The percentage of dealers who describe themselves as "active AI users" but cannot name a single measurable outcome from that use. Source: Cox Automotive AI Readiness Study.
Active use without measurement is the definition of the gap this newsletter exists to close.
WORTH READING
Cox Automotive AI Readiness Study — the primary data source behind the 34% figure and several other benchmarks we'll return to. Worth reading in full.
STELLA Automotive AI — their annual survey on AI adoption in dealerships has become one of the more reliable primary sources in the space.