Cognitive Offloading and the Bill That Comes Due An engineering manager who oversees teams using Claude Code and Codex argues that AI coding assistants create "comprehension debt" — working code that no one on the team can explain. The piece cites an Anthropic randomized trial in which professional developers learning an unfamiliar Python library scored about 17 percent lower on a comprehension quiz when assisted by AI, with the heaviest delegators averaging under 40 percent, as well as a Microsoft and Carnegie Mellon survey linking confidence in AI to reduced critical thinking. Every time I write a grocery list I am offloading mental effort. So is anyone who uses a calculator, a calendar, or Google Maps. Psychologists have a tidy name for this: cognitive offloading, which Risko and Gilbert https://doi.org/10.1016/j.tics.2016.07.002 defined back in 2016 as using physical action to reduce the mental demands of a task. It is one of the oldest tricks of our species. Writing itself is cognitive offloading. Nobody thinks the invention of the notebook made humanity dumber. So when people fret about AI making us stupid, my first instinct is to roll my eyes. My second instinct, after a year of managing engineering teams that use Claude Code and Codex all day, is to stop rolling my eyes and actually look at the research. It turns out the researchers have been busy, and the picture is more specific and more uncomfortable than the usual “kids these days” complaint. The key distinction, which I had not appreciated until recently, is between offloading a task and offloading the learning that the task would have produced. A grocery list offloads the task of remembering milk. It does not prevent you from knowing what milk is. But when you hand an LLM a problem you have never solved yourself, you offload the task and the learning at the same time, and you may not notice, because the task still gets done. The University of Technology Sydney put out a report https://sendfull.substack.com/p/ep-89-cognitive-offloading-to-ai this spring drawing exactly this line: offloading is fine until the thing you offload is the work that would have built or maintained a skill. At that point it stops being offloading and becomes atrophy. The evidence for this in software is now direct rather than anecdotal. Anthropic ran a randomized trial https://www.anthropic.com/research/AI-assistance-coding-skills in which professional developers learned an unfamiliar Python library, half with an AI assistant and half without. The AI group finished faster, as you would expect. Then everyone took a quiz on what they had just built. The AI group scored about 17 percent lower, and the people who had delegated the most to the model did the worst, averaging under 40 percent. The paper’s authors call this a tradeoff between productivity and skill formation, which is polite. What they observed was working code produced by people who could not explain it, which is a sentence I have now written in three different places because it keeps being the whole problem. That study was done on experienced engineers. A follow-up line of work https://arxiv.org/pdf/2604.04721 looking at persistence found that AI assistance also makes people give up sooner when the assistant is taken away, and the literature on novices is worse still, because a beginner has no prior expertise to compare the model’s output against. They cannot tell a good answer from a fluent one. Microsoft and Carnegie Mellon surveyed https://www.microsoft.com/en-us/research/publication/the-impact-of-generative-ai-on-critical-thinking-self-reported-reductions-in-cognitive-effort-and-confidence-effects-from-a-survey-of-knowledge-workers/ a few hundred knowledge workers last year and found that the more confidence someone had in the AI, the less critical thinking they reported doing. Nobody in that survey was being lazy. They were being reasonable. If the tool is usually right, checking it feels like a waste of time. That is precisely the trap Lisanne Bainbridge described in “Ironies of Automation”