Big-company AI is not the SMB playbook

Big-company AI is not the SMB playbook. It is the warning shot.

Reuters reported that Cloudflare plans to cut about 20 percent of its workforce, more than 1,100 jobs, while reorganizing around rapid AI adoption. The company had 5,156 full-time employees at the end of 2025 and expects $140 million to $150 million in restructuring charges. In its own note, Cloudflare said AI usage inside the company rose more than 600 percent in three months and that employees now run thousands of AI agent sessions each day across engineering, HR, finance, marketing, and other functions.

Microsoft points in the same direction from a different angle. GeekWire reported that Microsoft disclosed a $900 million charge for a one-time voluntary retirement program. The company said headcount declined year over year and is expected to decline again in fiscal 2027. At the same time, Microsoft plans more than $40 billion in current-quarter capital expenditures, mainly for data centers and AI infrastructure.

This is not only a tools story. It is an operating-model story.

A large company can look at agent-supported work and ask which roles should disappear, which teams should be rebuilt, which experienced people should be bought out, and which processes should be redesigned from the top down. Public companies have the capital, the management layers, and the market pressure to make that kind of change all at once. Whether any specific version proves wise is a separate question. The signal is already clear: AI is moving from individual productivity into company architecture.

Small businesses should not copy the layoff playbook. Most SMBs are not overbuilt. They are under-systematized.

An owner remembers the exception. A senior technician remembers the workaround. An office lead remembers the customer promise. An estimator knows which photos matter. A dispatcher knows which crew can handle the weird job. The business works because capable people carry private context that the operating system never captured.

AI does not fix that by itself. It exposes it.

Agents can summarize notes, draft replies, classify callbacks, prepare estimates, reconcile invoices, monitor reviews, and surface exceptions. A twenty-person company can become more disciplined than a fifty-person company used to be. Confusion can move faster too. Without a warranty standard, a discount-approval rule, a clean-closeout definition, and an owner for improving dispatch, the agent inherits the mess.

The E-Myth lesson becomes current again here. Michael Gerber’s old warning was that technicians do not become business owners just because they know the technical work. A technician knows how to do the job. An owner-operator knows how the job should be taught, measured, delegated, corrected, and improved.

AI raises the value of that second skill.

Future value will not sit only with the person who can do the task. It will move toward the person who can define the standard behind the task, supervise the agents that touch it, read the scoreboard, catch the exceptions, and improve the method.

Hadto’s thesis is to convert technicians and domain experts into owner-operators. Give them the workflow rules, scoreboards, customer promises, sales surfaces, apprenticeship loops, and authority to improve the system. Do not leave them as labor inside someone else’s machine, and do not hand them software that assumes the business model already exists.

This difference matters.

A technician with an AI tool is still exposed if the company above them decides the tool can absorb the role. An owner-operator with infrastructure has a different position. They can own the customer promise, the workflow, the measurement system, the agent boundaries, and the training path for the next person.

This is not anti-AI. It is the opposite. It treats AI as a force that makes ownership more important, not less.

Cloudflare and Microsoft show what big-company AI adoption can look like: fewer roles, retirement incentives, infrastructure spending, redesigned teams, and work design imposed from the top. The SMB answer should be different. Use the same pressure to build more owners.

The next business owner is not just the best technician in the room. It is the person who can make the room’s best work repeatable, measurable, and improvable without pretending the agents are in charge.