This is happening now. One of the top five drivers of developer burnout is not too much product work. It is poorly thought out AI initiatives being pushed from the top of the company without a real strategy.
The mandate is usually simple: AI has to be in the product, AI has to be in the workflow, and AI has to be visible in the metrics. Teams are told to add chatbots whether or not they help users, report monthly and weekly movement in whatever numbers leadership wants this quarter, and keep defending the initiative even when the goals keep changing.
That is exhausting. Developers are not just building software anymore. They are absorbing the cleanup from an executive program that keeps changing shape.
Burnout starts with moving targets
A poorly thought out AI initiative usually creates more reporting than relief.
The team is asked to track monthly improvements, then weekly improvements, then some new adoption metric that replaces the last one because leadership wants a better story for the next review. Nobody agrees on whether the goal is speed, cost reduction, higher conversion, fewer tickets, or better retention. So the developers end up supporting a program that is always being measured and rarely being understood.
That gets worse when leadership demands chatbots be installed in products regardless of whether they help users. The chatbot becomes a checkbox. It is forced into onboarding, support, search, settings, sales flows, and internal tools because the company wants visible AI, not useful AI. Engineering has to wire it up, maintain it, handle failures, and explain why the feature creates more friction than value.
So the team is not just building software. It is defending software that should not have been shipped in the first place.
The contradiction at the top
This pressure usually comes from the C-level because the message is inconsistent from the start.
On one hand, AI is non-negotiable. On the other hand, every request for spend is scrutinized down to the last twenty dollars. Leadership insists the initiative is strategic, but then treats the project like a line-item test of discipline. The result is predictable: the company demands speed, visibility, and transformation, while refusing to fund the experimentation, cleanup, and operational overhead that real AI systems require.
That contradiction gets handed to engineering.
Developers are asked to justify every API call, every model invocation, every internal tool, every support burden, and every half-working prototype, even though the mandate itself came from above. They are told to move fast, but also to prove value immediately. They are told AI is essential, but also that every small expense must be defended like a mistake.
That is not leadership. It is pressure transfer.
And pressure transfer burns people out.
What a real strategy would look like
A real AI strategy would start smaller and stay honest.
It would not ask for chatbots in every product by default. It would ask where AI actually removes work, reduces risk, or improves the customer experience enough to matter. It would define success before deployment. It would keep the metric set stable long enough to learn something. It would fund the work at a level that matches the ambition instead of pretending transformation can happen on a skeptical budget and a moving target.
Most importantly, it would stop making developers clean up executive theater.
If AI is truly strategic, then it should be used selectively, measured carefully, and supported properly. Otherwise the company is not building leverage. It is turning engineering into the shock absorber for leadership’s latest mandate.
Get in touch if your AI program is creating burnout faster than it is creating value.