In a Business Insider personal essay, software engineer Siddhant Khare writes about what he calls “AI fatigue” — a mental exhaustion that comes not from AI tools being hard to use, but from their relentless integration into everyday coding work. Khare explains that while AI has dramatically boosted his productivity (he shipped more code than ever before), it also made his job mentally harder because he’s constantly switching between multiple problems and reviewing AI outputs instead of doing deep, focused design work. This kind of frequent context-switching — juggling as many as six tasks a day — is “brutally expensive for the human brain,” he says.
Khare isn’t against AI itself — he actually builds AI tools — but he highlights a paradox of productivity: although AI reduces the cost of producing code, it increases the demand for coordination, review, and decision-making, which falls entirely on engineers. Because AI doesn’t tire, workers often feel pushed to take on more tasks and keep up with rapid tool updates from companies like OpenAI and Anthropic. This “fear of missing out” on the latest capabilities can push developers to spend weekends evaluating tools, reading changelogs, and following every demo.
Khare also worries about skill atrophy — that relying too much on AI can erode core problem-solving abilities. He compares it to GPS weakening our natural navigation skills; similarly, he finds that his capacity to reason through algorithms or concurrency problems without a laptop has declined. Former Tesla AI lead Andrej Karpathy has echoed similar observations, noting that advanced AI coding tools can slowly dull developers’ manual coding skills.
To cope, Khare suggests setting personal boundaries around AI use and taking intentional breaks from the AI conversation (for example, ignoring updates during his holiday). He also calls on AI companies to build human-centric guardrails that help prevent cognitive overload and burnout — warning that without such safeguards, engineers risk wearing down even as AI makes them more “productive.”
