Last week I had coffee with a senior executive at one of Europe’s leading enterprise software providers. We talked about my earlier article on the dot-com lessons for AI. In that article I described how we burned €28 million on e-commerce hype, because we started with the technology instead of the customer problem.
He recognized every pattern because he was there also. Then he showed me a Guardian article about SemiCab, an AI-powered freight platform from Algorhythm Holdings. “Same story, different decade,” he said. “And this time, the market overreacted in both directions on the same day.”
He was right.
What happened
Algorhythm announced that its SemiCab platform lets operators scale freight volumes by 300% to 400% without adding staff. In live deployments, individual operators manage over 2,000 loads per year. The industry average is roughly 500. The platform also cuts empty trucking miles by 70%. On paper, impressive numbers.
The stock market reacted fast. C.H. Robinson dropped 14.5%. RXO lost over 20%. J.B. Hunt, XPO, and Expeditors all fell hard. In one trading session, billions in market value disappeared from companies that move the world’s freight every day.
Two extreme reactions. Excitement about the disruptor. Fear about the incumbents. In my view, both were wrong.
The technology is real. The question is momentum.
The underlying logic of SemiCab makes sense. Trucks drive empty one out of every three miles. That is a huge and well-documented waste. Using AI to plan collaborative round trips across a network of shippers and carriers is a classic optimization problem. The math works.
This is the kind of non-generative AI I described in my earlier article. Not flashy. Not generating content. But embedded in operational decisions. The EDI of our time, not the website.
And here is the point many people miss: Working technology is necessary but it is not enough.
What matters just as much is fulfilling the preconditions that turn a proven concept into a market reality.
Market adoption. Are enough shippers, carriers, and logistics providers ready to join a collaborative network? Collaboration in freight sounds simple. But it requires trust, data sharing, and a willingness to change how business has been done for decades. That takes time.
Scalability. SemiCab’s results come from deployments in India. The US expansion through their Apex SaaS platform is just starting. Scaling across different regulatory environments, carrier structures, and commercial cultures is a very different challenge than proving the concept in one market.
Expandability. Can the platform grow beyond its current customer base? Can it handle multi-modal, multi-country logistics? These are open questions.
Network effects. The company itself says that results improve as more participants join the network. That is the upside. But below a critical mass of adoption, the value may not be strong enough to change behavior.
I have seen this before
In 2000, I joined a global e-commerce initiative at a traditional building materials company. The technology worked. SAP Internet Transaction Server was functional. B2B marketplaces were technically sound. The vision of frictionless digital procurement was compelling.
The problem was not the idea. And not the technology. The market was simply not ready.
Customers did not want to change how they bought. Internal systems did not connect. Governance across stakeholders was slow. Relationships still mattered more than price transparency. Two years and €28 million later, the program was scaled back.
E-commerce did not disappear. It did transform B2B over time. But the adoption curve and organizational readiness lagged the hype by at least ten years.
I see the same pattern here.
Why selling incumbent stocks is just as wrong
The sell-off in logistics stocks shows a market that confuses possible disruption with immediate disruption.
C.H. Robinson, XPO, and J.B. Hunt have massive physical networks. They have deep customer relationships, decades of operational experience, and established trust. The idea that one AI platform announcement wipes out that value overnight is the same thinking that inflated dot-com valuations in 1999.
Will AI change freight brokerage? Yes. Will companies that do not adapt lose ground? Yes. But the shift from manual, relationship-driven freight management to AI-run networks will take years. Not quarters. And incumbents that adopt AI themselves, and embed it into their existing networks and customer base, may end up stronger.
The market priced in a revolution. What is happening is the beginning of an evolution.
The real lesson: preconditions come first
This brings me back to something I see in every AI discussion with executives. People focus on what the technology can do, ignoring real life is different from pilot environments. The harder question gets less attention: what needs to be true for the technology to work at scale?
In my earlier article I described five dimensions that determine transformation success: People, Process, Technology, Data, and Policy. SemiCab scores well on one of those five. But for the freight industry to change the way the press releases suggest, all five need to be in place.
You need carriers willing to share data. You need shippers willing to change how they buy transport. You need commercial rules that reward collaboration over individual gain. You need clean, standardized data across thousands of participants. And you need people who trust the system enough to let go of manual control.
That is the hard work. A whitepaper does not make it happen.
What should executives take from this?
The SemiCab story is worth following. The technology targets a real and large problem. The potential value is there. But potential is not the same as proof at scale.
If you run a logistics business, do not panic. Ask the questions that matter instead. Where do our customers feel pain? What decisions could AI make better? Is our data ready? Do we have the governance to make changes last?
If you invest in this sector, do not treat every AI announcement as a reason to sell everything. Look at adoption. Look at what needs to change beyond the technology before value shows up.
And if you sit on a board evaluating AI plans, remember this: the companies that won after dot-com were not the ones with the best technology. They were the ones that connected technology to a business model that served customers and made money.
The opportunity is real. The timeline is longer than the headlines suggest. And it is not about the technology. It is about what you deliver to customers, and whether you can keep delivering it.
That was true in 2000. It is true today.
Sikko Zoer is Senior Partner at Qwinn Business Partners, a boutique supply chain consulting firm. This article builds on his earlier piece We Burned €28 Million on Tech Hype. Here’s Why Your AI Budget is Next.
