Tech Workers Who Maxed Out on AI Are Now Pulling Back

A quiet counter-movement is emerging among tech workers who were among the earliest and heaviest adopters of AI productivity tools: having pushed AI use to its limits, they are now consciously trying to reduce it. The Indian Express reports on a growing cohort of developers, writers, and knowledge workers who found that maximising their use of tools like ChatGPT and GitHub Copilot came with unexpected downsides — atrophied problem-solving skills, over-reliance on outputs they couldn't always verify, and work that felt increasingly disconnected from their own judgment.

The phenomenon is distinct from AI scepticism. These workers are not anti-AI; most continue to use the tools selectively. Instead, they describe a deliberate recalibration — reserving AI assistance for specific tasks while deliberately working through others unaided, in much the way a runner might choose not to use a GPS watch to rebuild their sense of pace. The goal is to preserve the underlying skill even while using the tool.

The backlash points to a broader tension in how AI is being integrated into professional work. Employers and productivity evangelists have pushed maximum AI adoption as a competitive imperative, but workers on the ground are discovering that the relationship between AI assistance and human capability is more complex than a simple productivity multiplier. Skill erosion — the gradual loss of competence through disuse — is emerging as a real concern, particularly among junior workers who may never have developed certain skills in the first place.

For India, where millions of IT professionals are navigating rapid AI-driven change in their workplaces, the debate over how much AI use is too much carries significant stakes. The workers pulling back may be early signals of a wider industry conversation about the right balance between AI leverage and human skill development.