Lauren Tuffet, Billionaire

When Lauren Tuffet died in 2071, the obituaries credited her with seeing the Planetary Body before almost anyone else did.

By then, the phrase had entered ordinary language. It described the sensors, networks, controls, machines, power systems, security layers, and maintenance infrastructure through which artificial intelligence had acquired a physical presence in the world.

In 2026, attention was still fixed on the brain. Technology companies were spending extraordinary sums on chips, models, and data centers. Investors studied benchmark scores, while reporters covered each model release as though a new species had appeared.

Tuffet wondered what a brain would need next.

AI knew a great deal about the recorded world, but almost nothing about what was happening at a particular place at a particular moment. It could explain how bridges fail without knowing that one was beginning to fail. It could identify water damage in a photograph without knowing that rain had entered a particular wall the night before. It could diagnose a machine from vibration data, provided someone had installed a device capable of feeling the vibration.

Sensors seemed like the beginning. Cameras, microphones, moisture probes, vibration monitors, medical devices, satellites, and industrial instruments could give AI access to present conditions. Those signals would need to travel, and some decisions would need to be made locally. Intelligence would also need ways to move vehicles, open valves, stop machines, and direct people. Power, cooling, security, maintenance, and operational memory would follow.

Tuffet called the emerging system the Planetary Body.

Her first mistake was assuming that the largest categories would produce the largest fortunes. She became briefly fascinated with sensor manufacturers, reasoning that trillions of devices might eventually be installed. A year of research persuaded her that many of those devices would become cheap, interchangeable, and difficult to distinguish.

During a visit to a food-processing plant, an engineer showed her a sensor that cost less than dinner for two. Replacing it required a shutdown, a lift, two technicians, protective equipment, and several hours of paperwork. The device was nearly free; reaching it was expensive.

After that, Tuffet developed an irritating habit of asking people what the least impressive part of their system was that could stop everything.

At an automated warehouse, it was a dirty optical window. At a municipal facility, it was an obsolete controller that nobody wanted to replace because the conversion would interrupt operations for several days. A manufacturer experimenting with predictive maintenance discovered that six acquired plants used different names for the same class of pump.

These were minor facts in an era preoccupied with general intelligence. Tuffet found them more useful than another prediction about which model would win.

She was hardly immune to fashion. Early notes show that she spent too much time on humanoid robots and held onto one sensor company long after its service costs had begun swallowing its margins. Her later reputation for patience concealed a stubborn streak that occasionally looked much the same from the outside.

Still, she kept returning to the awkward places where intelligence met machinery, electricity, weather, regulation, maintenance schedules, and human reluctance. Over time, she became less interested in the size of a market than in the junctions the market could not avoid.

Decades later, admirers assumed that the Planetary Body thesis had originated with a private research team or a network of industrial experts. Among the files released with Tuffet’s papers was an exported chat dated July 2026. The prompt was intact:

Artificial intelligence is developing as a brain before it has developed a complete physical body.

Build a serious technological and economic thesis around the idea that the next major phase of AI will require a planetary-scale body through which intelligence can perceive, communicate, decide locally, act physically, obtain power, defend itself, preserve memory, and remain operational.

Begin with AI models and compute as the brain. Treat literal physical sensors as the receptors through which AI gains direct contact with present reality.

From there, identify only the additional technological systems that are functionally unavoidable. Do not create a childish one-to-one list of human organs. Use the organism metaphor only where it reveals a real dependency, bottleneck, or economic layer.

For each layer, address the following six points:

Explain why increasingly capable AI makes the layer necessary.

Identify the technologies, industries, infrastructure, and services that supply it.

Distinguish expected growth in usage from the ability of businesses to capture durable economic value.

Explain where commoditization is likely and where scarcity, switching costs, certification, installed-base advantages, recurring revenue, or technical difficulty may remain.

Identify second- and third-order dependencies that are easy to overlook.

Explain what would have to be true for the thesis to fail.

Across the thesis, pay particular attention to sensors, connectivity, edge computing, industrial controls, robotics and actuation, power generation and distribution, cooling, cybersecurity, device identity, operational software, calibration, maintenance, repair, and retrofit of existing physical infrastructure.

Do not recommend stocks, securities, or investments. Do not assume that a rapidly growing market is necessarily a profitable one.

Produce a dependency map showing how intelligence moves from computation, to perception, to judgment, to physical action, and finally to verified consequences.

End by identifying the small number of indispensable junctions through which most plausible versions of the AI future would have to pass.

The answer was impressive, technical, and dangerously confident. It identified real bottlenecks, invented a few others, and presented both in the same authoritative voice. Tuffet had been given a useful map with several mislabeled roads.

When the prompt became public, millions of people ran it. Their models produced sector maps, market forecasts, company lists, and confident predictions about the machinery that would surround AI. Most people read the answer, found it interesting, and closed the window.

Lauren Tuffet went looking for evidence.