With all the discussion about AI agents lately, you might be wondering: how exactly do you count them? If multiple agents collaborate to resolve a customer issue or approve a loan application, does that represent one digital worker or many? The question may sound trivial, but it will soon matter a great deal. Organizations will eventually track digital headcount the same way they track human employees today.

A dangerous new market narrative is spreading through boardrooms and earnings calls: artificial intelligence has made companies so productive that they can slash their workforce and barely notice the difference. Analysts applaud, the stock jumps, and executives describe a future where digital labor replaces the old human-heavy operating model. Unfortunately, the economy is rarely that tidy.

There is, perhaps, a small silver lining in the current wave of AI anxiety. Not long ago, the dominant fears revolved around killer robots, runaway superintelligence, and apocalyptic scenarios that ended with data centers being nuked from space. Today the panic is more grounded, and in many ways more sophisticated. We are no longer imagining machines conquering humanity; we are worrying about white-collar unemployment ticking above 10%, mortgage books wobbling in San Francisco, and private credit portfolios unraveling because software agents can write code faster than junior analysts. The monsters have moved from science fiction to the balance sheet.

The AI crisis arrived without fanfare. There were no alarms, no cascading red dashboards, no breathless messages from the security operations center. At 3:17 a.m., somewhere between Singapore and Rotterdam, an AI routing agent inside a global logistics company glitched slightly. Just milliseconds. But that agent sat at the center of thousands of shipments, negotiating contracts, rerouting containers, balancing fuel costs and port congestion in real time. That week, the company cloud provider had quietly shifted workloads to a different region after an energy price spike. A frontier model vendor had rolled out an update that subtly changed how the system reasoned. An external technology partner, granted limited access months earlier, had folded usage patterns into broader product improvements now available to competitors. Nothing was hacked. Nothing was stolen. Yet by quarter’s end, delivery times slipped, margins thinned, and the firm’s once sharp operational instincts felt strangely generic.

In the mid nineties, designers did not know what online shopping was supposed to look like. So they borrowed from the physical world. Early retail websites featured isometric shopping carts gliding down digital aisles. Shelves were rendered in crude 3D. You clicked arrows to “walk” through a store. After a while, search bars replaced aisles, and recommendation engines became the new merchandising layer. Eventually, mobile screens collapsed the store into a feed. We are now at another such inflection point. Retailers are redesigning the storefront again. But this time, the shopper may not even be human.