Every few years, a new word arrives in the automotive industry wearing the same costume: the future.
It shows up at conferences first, usually attached to a slide deck with a bold font and a stock photo of someone staring meaningfully at a glowing screen. It fills trade publications. It dominates roundtables. Vendors build their entire pitch around it. Consultants write white papers. Dealers get emails with subject lines like "Is Your Store Ready for [Buzzword]?"
Then, quietly, it disappears.
Not all at once. There's no funeral. Just a slow retreat from the keynotes, a gradual fading from the sales pitches, until eventually you stop hearing it and you realize you can't remember the last time you did.
I’ve lived through a lot of these cycles in the automotive industry. Looking back, they all seemed different at the time. Today, many of them share something important in common.
Understanding that common thread explains why AI is different. The reason isn't philosophical.
It's mechanical.
Blockchain was going to make title fraud extinct. Every vehicle's history, every service record, every ownership transfer, and every accident would be immutably logged on a distributed ledger that no one could tamper with. Auctions would become transparent. Dealers would save hours of paperwork.
It wasn't just a compelling vision. We explored putting it into practice ourselves.
In 2021, we evaluated blockchain for an international marketplace. The concept was solid. We needed seven organizations connected to create end-to-end accountability across the supply chain. We secured five. Two never came on board.
The technology worked. The vision made sense. But the value proposition depended on all seven organizations participating. Missing just two critical participants was enough to unravel the entire model.
Most dealers never touched blockchain directly. Not because blockchain lacked technical merit.
It struggled because it required an entire ecosystem to change simultaneously. Dealers, OEMs, lenders, insurers, auctions, repair facilities, DMVs, software providers, and consumers all had to participate before its advantages became meaningful.
That's an extraordinarily high adoption hurdle, especially when the existing system, while imperfect, already functioned well enough for most participants.
Looking back, that's the pattern I see repeated across many of automotive's biggest technology waves.
The question was rarely whether the technology worked. The question was who had to change before it became valuable.
If the answer was "everyone," adoption was always going to be slow.
NFTs and the metaverse followed a remarkably similar path. Automakers experimented with digital collectibles, ownership certificates, loyalty programs, and immersive virtual showrooms where customers could configure vehicles, interact with a salesperson's avatar, and complete an entire purchase in a digital world. The vision was compelling, and the marketing was everywhere. But consumer behavior never shifted as dramatically as the presentations predicted.
The NFT market collapsed, enthusiasm for the metaverse cooled, and most buyers still preferred to walk around, sit in, and drive the actual vehicle before making one of the largest purchases of their lives.
Autonomous vehicles may be the most instructive example. For nearly a decade, the dominant narrative was that self-driving cars were always just around the corner. Eighteen months away. Then twelve. Then two years. Then subject to regulatory review. The technology is genuinely remarkable, but the gap between "works in controlled testing" and "works safely in every weather condition, on every road type, in every city" proved far larger than anyone anticipated.
The promise hasn't disappeared. It's been recalibrated. Which is different.
Digital retailing followed a different path.
Unlike blockchain or the metaverse, digital retailing solved meaningful problems almost immediately. Customers could value their trade online, estimate payments, begin financing, reserve inventory, and reduce time spent in the showroom. Dealers improved efficiency, generated better-qualified leads, and created a more convenient buying experience.
Most dealers invested, and many continue investing today.
Customer adoption has grown, but the fully online, fully touchless buying experience has proven to be a preference held by a much smaller percentage of buyers than the hype implied.
That's not because the platforms failed. In fact, many of today's digital retailing solutions are remarkably sophisticated, and we've had the opportunity to work alongside several of them. The challenge is that buying a vehicle isn't a single workflow. It's dozens of interconnected workflows.
Inventory. Pricing. OEM incentives. Credit applications. Identity verification. Trade valuations. Lender integrations. F&I products. Taxes. DMV requirements. Title and registration. Document generation. Electronic signatures. Payment processing. Delivery scheduling.
Each may involve a different provider, different APIs, different business rules, and different release cycles.
Building an experience that appears seamless requires integrating an extraordinary number of independent systems. Building those integrations is expensive. Maintaining them is often even more expensive as providers evolve their APIs, OEM programs change, regulations shift, and vendors continuously modify their products.
It's one of the reasons so many digital retailing platforms excel at portions of the buying journey, but relatively few deliver a truly end-to-end experience. The technology exists. The challenge has always been making dozens of independent technologies behave like one cohesive product while still delivering a return on the investment required to build and maintain it.
It wasn't simply a case of the industry resisting change.
It was another reminder that solving a complex business problem is very different from building an impressive product demonstration.
Then came AI.
Unlike the technologies that came before it, AI doesn't require the rest of the industry to move first.
That's the fundamental difference.
Blockchain became more valuable as more organizations joined the network.
AI becomes valuable the moment one person starts using it.
One engineer writes code faster.
One designer explores more ideas.
One product manager analyzes requirements more efficiently.
One dealership employee summarizes customer feedback in minutes instead of hours.
No ecosystem-wide rollout.
No industry standards.
No waiting for everyone else to catch up.
The return on investment starts on day one.
At Trupropel, AI is part of how we build products every day. More importantly, it's something we've deployed for our clients, not just experimented with internally. We recently embedded an AI agent inside a client's marketing data platform, and what we learned building it reinforced the very point this article makes: AI creates value when it's paired with the experience to ask the right questions, make the right decisions, and know when the first answer isn't the right one.
That's also why we're skeptical of "AI-powered" labels. The model is only one part of the solution. The real advantage comes from the product, design, and engineering experience that surrounds it.
Ironically, using AI every day has also made us more skeptical of AI marketing.
About eighteen months ago, we were discussing a dashboard project with a prospective client. During discovery, they asked us to incorporate AI into the product.
After reviewing the requirements, we gave them an answer they probably weren't expecting.
"You don't actually need AI."
The application already solved the problem well. AI wasn't necessary to deliver the functionality they wanted.
Their response was refreshingly candid.
"We know."
"But if we include AI, we can charge another $199 per month."
They were prepared to invest nearly $50,000 in additional development because they estimated the increased subscription revenue would recover that investment in roughly four months. After that, it would become almost entirely incremental margin.
If they were going to market AI, they wanted customers to receive genuine AI functionality rather than simply adding an "AI Powered" badge to their website. Their decision wasn't driven by engineering.
It was driven by customer expectations and commercial reality.
We left that conversation with a simple test we've used ever since: If you removed the AI tomorrow, would your customers lose anything, or would only the marketing change?
That conversation has stayed with me because it perfectly captures where the industry finds itself today.
AI is simultaneously becoming one of the most valuable engineering tools we've ever adopted and one of the most powerful marketing terms the software industry has ever produced.
Sometimes those two overlap.
Sometimes they don't.
Having spent more than four decades in automotive, from auction and wholesale operations to CarMax, Lithia, co-founding The Appraisal Lane, and now building software products for dealers, auctions, lenders, and OEMs, I've had the opportunity to watch these technology waves arrive from several different vantage points.
I've seen genuinely innovative products struggle because they required too many organizations to move together.
I've also seen relatively simple ideas spread rapidly because they created value for one person on the very first day.
The pattern across all of these technologies is surprisingly consistent. The technology itself was rarely the problem. The mistake was assuming that technical capability automatically creates market adoption. History suggests otherwise.
A technology becomes valuable not because it's new, but because it solves a problem that someone feels strongly enough about to change their behavior, improve their business, or justify paying for it.
The automotive industry, with its long product cycles, relationship-driven sales culture, regulatory complexity, and enormous installed infrastructure, is particularly resistant to pure technological disruption. Not because the people in it resist change, but because the problems are genuinely hard and the switching costs are genuinely high.
Every generation has its "next big thing."
The technologies that survive aren't the ones with the loudest marketing.
They're the ones that quietly become indispensable.
They become woven into daily operations and established workflows. They improve quality, reduce friction, create measurable value, and eventually disappear into the background because using them simply becomes the normal way of working.
Looking back over the last forty years, that's the real test.
Not whether a technology can generate headlines.
Not whether it represents a remarkable technical achievement.
The technologies that endure are the ones that create value for someone today, without requiring everyone else to change tomorrow.
That's why I believe AI is different.
The hype will fade.
The utility won't.
