--- title: Reducing the cognitive load of AI changes | daily.dev description: A developer shares a technique for reducing the mental overhead of reviewing AI-generated code: before reviewing changes, they prompt the AI to extract any... canonical: https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Reducing the cognitive load of AI changes | daily.dev og:description: A developer shares a technique for reducing the mental overhead of reviewing AI-generated code: before reviewing changes, they prompt the AI to extract any... og:url: https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf og:image: https://api.daily.dev/og/posts/rvZpZjKAF.png og:image:alt: Reducing the cognitive load of AI changes og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Reducing the cognitive load of AI changes **[Lobsters](https://daily.dev/sources/lobsters)** · 2 min read · 0 upvotes · 0 comments ## Summary A developer shares a technique for reducing the mental overhead of reviewing AI-generated code: before reviewing changes, they prompt the AI to extract any unconventional or bespoke naming it introduced for abstractions, list meanings and alternatives in a markdown file, then confirm or override each term so the AI can find-and-replace it everywhere, including documentation. The result is code that reads as if it used the reviewer's own vocabulary, making review significantly easier. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I make AI-generated code easier to review when the AI picks confusing names for things? Prompt the AI, before you review, to extract any unconventional or bespoke terms it used for abstractions, processes, and concepts into a markdown file listing each term's meaning, why it was chosen, and alternative options. Confirm or replace each term, then have the AI find-and-replace the chosen terms everywhere, including documentation, so the code matches your own mental vocabulary. _Developers refining their AI code-review workflow can find more practical prompting tricks like this on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 2 discussions and 8 comments across lobsters, hackernews (as of 2026-10-02). **TL;DR:** Developers broadly welcome the idea of keeping a shared glossary to tame AI-invented terminology, noting that models tend to coin new and often worse names more freely than humans. **Sentiment:** 40% positive · 55% mixed · 5% skeptical **The case for** - A glossary helps not just naming but overall code quality by clarifying each component's responsibility. - Shared vocabulary is already a well-established good practice (e.g. Domain Driven Design), so this just adapts it to AI workflows. - Renaming AI-generated terms to be more precise (e.g. distinguishing estimates from real data) can help reinforce correct semantics in the code. **The pushback** - Models seem to invent confusing new names even more freely than humans, and the names are often worse. - AI appears to pull in terminology indiscriminately from many different codebases, like a new hire with no consistent background. - Humans are forced to be more 'semantically efficient' with naming due to limited cognitive capacity, which may make human-written code better by default. **By community** - lobsters (positive): Commenters generally endorse glossary-style approaches and share related experiences taming AI-invented terminology. - hackernews (mixed): No substantive comments were provided to characterize the discussion. **Open questions** - Whether AI's tendency to invent new terminology stems from hidden chain-of-thought reasoning not reflected in its visible output. - Whether forcing more precise naming (e.g. distinguishing 'estimate' from 'actual') actually changes a model's underlying confidence or behavior. **Highlights** > Having a shared vocabulary of concepts is essential for all programming, in no way limited to AI. (It’s the key feature of Domain Driven Design, for example.) But LLMs tend to invent new confusing names even more freely than humans, and the names tend to be worse! So anything you can do to rein in the naming conventions is helpful. I just use a glossary section of the design overview document. > — [wrs on lobsters · 2 points, 1 comments](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes#c_uc1zav) > AIs are kind of a mishmash of every codebase out there. I think their terms are lifted from somewhere far away rather than made up. It's like working with someone who just started at the company. Every day, and their last job was every job. > — [landon on lobsters · 2 points, 1 comments](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes#c_gkvyv4) > I vibed a small android app for doing workout tracking. It includes a weight estimator component for figuring out cross lift equivalences. In the process, the AI became much more confident that the estimator and the simulator for it were more accurate than the actual data from my work outs. I ended up doing a big renaming pass to force it away from using mathematical names for things and to have names that explicitly included "estimate" like "estimated1RM" or "bicepStrengthEstimate". Too soon to say whether that had the desired effect of reinforcing where truth versus estimate comes from, but I wouldn't be surprised if the math formalisms nudged its weights towards more certainty then it should have had. > — [mfk on lobsters · 2 points](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes#c_gif5ul) > I strongly agree that this is a problem with current AI systems. My intuition is that the AI has a much greater patience for new terms. Us humans, with our comparatively limited capacity to juggle concepts, are forced to be more efficient with which words we choose to pull in to our code. The result is we still write better code. We're more "semantically efficient". > — [3lambda on lobsters · 1 points, 1 comments](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes#c_oroec5) > I suspect that some part of this is the hidden reasoning that models do these days. You don't see any of the chain of thought stuff going on prior to the model's output, only the polished output. So there's some assumption baked into the weights that the context that you're dealing with when reading has a bit of an understanding of the CoT. That feels like it would be a difficult thing to train out of a model as it's effectively use this info to work out what to say, but don't actually reference it. It's like a game of charades happening somewhat. I worked at a frontier lab, but not on anything model side, so this is just a fairly lay understanding of this, I could be very wrong. > — [joshka on lobsters · 1 points](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes#c_8dnybs) **Source threads** - [lobsters](https://lobste.rs/s/iwvjr0/reducing_cognitive_load_ai_changes) · 30 points · 8 comments - [hackernews](https://news.ycombinator.com/item?id=49927109) · 2 points · 0 comments ## Similar posts on daily.dev - [Slowing Down AI On Purpose](https://daily.dev/posts/slowing-down-ai-on-purpose-romxvvrsp) · RUBYLAND · 1 upvotes · 0 comments - [AI Code Review Best Practices](https://daily.dev/posts/ai-code-review-best-practices-ii7hbaa3j) · Aviator Blog · 15 upvotes · 1 comments - [I rarely use AI to write code, but here's how it improves my coding workflow](https://daily.dev/posts/i-rarely-use-ai-to-write-code-but-here-s-how-it-improves-my-coding-workflow-jyjtuxvvk) · XDA Developers · 2 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-coding](https://daily.dev/tags/ai-coding), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Reducing the cognitive load of AI changes","url":"https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf"},"datePublished":"2026-10-01T21:18:39.759Z","dateModified":"2026-10-02T02:37:23.349Z","description":"A developer shares a technique for reducing the mental overhead of reviewing AI-generated code: before reviewing changes, they prompt the AI to extract any...","image":"https://media.daily.dev/image/upload/s--foaA6JGU--/f_auto/v1722860399/public/Placeholder%2004","thumbnailUrl":"https://media.daily.dev/image/upload/s--foaA6JGU--/f_auto/v1722860399/public/Placeholder%2004","isAccessibleForFree":true,"articleSection":"Lobsters","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Lobsters","logo":"https://media.daily.dev/image/upload/s--tl8v_Fku--/f_auto,t_logo/v1698841318/logos/lobste.jpg","url":"https://daily.dev/sources/lobsters"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ai-coding,prompt-engineering","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Lobsters","item":"https://daily.dev/sources/lobsters"},{"@type":"ListItem","position":3,"name":"Reducing the cognitive load of AI changes"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/reducing-the-cognitive-load-of-ai-changes-rvzpzjkaf#faq","mainEntity":[{"@type":"Question","name":"How can I make AI-generated code easier to review when the AI picks confusing names for things?","acceptedAnswer":{"@type":"Answer","text":"Prompt the AI, before you review, to extract any unconventional or bespoke terms it used for abstractions, processes, and concepts into a markdown file listing each term's meaning, why it was chosen, and alternative options. Confirm or replace each term, then have the AI find-and-replace the chosen terms everywhere, including documentation, so the code matches your own mental vocabulary. Developers refining their AI code-review workflow can find more practical prompting tricks like this on daily.dev."}}]} ```