--- title: What happens when the information runs out | daily.dev description: An exploration of AI colorization applied to old family photographs reveals a fundamental limitation: when source information is missing, AI fills gaps by... canonical: https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: What happens when the information runs out | daily.dev og:description: An exploration of AI colorization applied to old family photographs reveals a fundamental limitation: when source information is missing, AI fills gaps by... og:url: https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac og:image: https://api.daily.dev/og/posts/gZGDzl6ac.png og:image:alt: What happens when the information runs out 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. # What happens when the information runs out **[Hacker News](https://daily.dev/sources/hn)** · 6 min read · 0 upvotes · 0 comments ## Summary An exploration of AI colorization applied to old family photographs reveals a fundamental limitation: when source information is missing, AI fills gaps by inventing details — like eyebrows that were never on the negative. Working through several images, the author categorizes information into three types: known and describable, reasonably inferable, and simply lost. The key insight is that the human's role in AI-assisted restoration isn't just supervision but clearly communicating both certainty and uncertainty — knowing when a gap can be plausibly filled versus when it must remain permanent. The iteration process mirrors traditional darkroom work, where repeated refinement is the method, not a workaround. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Community take How the wider developer community reacted, aggregated from 1 discussion and 36 comments across hackernews (as of 2026-07-24). **TL;DR:** Commenters largely agree that AI image restoration fundamentally cannot recover information that was never recorded — it can only hallucinate plausible-looking details — and that the author's expectations were misplaced. A minority note nuances around deblurring, multi-frame video, and domain-specific tools. **Sentiment:** 10% positive · 40% mixed · 50% skeptical **The case for** - Multi-frame video contains delta information that can sometimes help recover detail beyond what a single frame holds. - Domain-specific deblurring (e.g., recovering blurred text by matching known fonts) can legitimately reconstruct absent information in constrained cases. - Representing uncertainty explicitly in data systems is a valid and underappreciated design principle. **The pushback** - AI restoration tools simply invent missing data rather than recovering it, making the author's expectation of faithful restoration fundamentally impossible. - Using a general-purpose tool like ChatGPT's image generator for precise restoration is the wrong tool for the job. - The author's instruction to 'not resolve details that didn't exist' is incoherent — upscaling/restoration inherently requires inventing details. - Information theory basics suggest blurring discards high-frequency data that cannot be recovered, and a video contains less information than the sum of its frames. **By community** - hackernews (mixed): Commenters are split between critiquing the author's naive expectations of AI restoration and offering nuanced technical caveats about when partial information recovery is genuinely possible. **Hottest debate:** Whether multi-frame video or deblurring operations can legitimately recover absent high-frequency information, or whether all such enhancement is fundamentally fabrication. **Open questions** - Are there purpose-built (non-ChatGPT) AI restoration pipelines that could do better at preserving fidelity while still upscaling? - How should users communicate the extent of AI-invented detail when presenting 'restored' images as historical artifacts? **Highlights** > 2 main problems with this article: one it's fundamentally asking for something impossible, and two for the part that _is_ possible it's using the wrong tool for the job. It's not possible to magic details into existence that weren't there. So if you think a color was wrong, well, that is not something that can be solved other than by using a time machine or finding a separate reference picture. Any human restoration artist would be doing the exact same thing of guessing the color. The second problem brought up is that ChatGPT is willing non-color details into existence that aren't in the original picture (eyebrows etc). For this it's simply that the public ChatGPT image generator wasn't made for this task. You'd have to build a custom harness but I'm sure it can be done. > — [Anon1096 on hackernews](https://news.ycombinator.com/item?id=49026498) > > but a series of frames can (I'm pretty sure) between them have enough info to deblur an object in the background You're probably thinking of geometrical super-resolution, such as gigapixel photography. Blurring is the discarding of high-frequency information, and only a very very small part of this information can be recovered using techniques like this (as in, so little that you wouldn't be able to notice it). > My only point is information in a video is more than the sum of the information of it's frames, It's actually less than the sum of the information of it's frames. An off-the-shelf lossless compression algorithm can give you an upper bound for the amount of information present in a video file. > — [wizzwizz4 on hackernews](https://news.ycombinator.com/item?id=49026011) > > is just going to be making up the missing data. This is generally true. What I find fascinating is the application of information that is not present in the data but that can be confirmed by the data. I’m thinking specifically of deblurring algorithms people use to recover text. If you can guess what font is being used in a screengrab with, say, someone’s bank account blurred out, you can take the i formation contained in the downloaded font and try combinations until you match the blur algorithm that may also be information not contained in the image. What’s also crazy, is that even if all you do is black out the text, kerning alone can still be used to guess the correct text with some degree of accuracy. I worked on redaction software at one point, and these were very much the types of things we thought about when dealing with text. > — [milesvp on hackernews · 1 comments](https://news.ycombinator.com/item?id=49026459) > I feel like the basics of information theory are a valuable thing for people to learn, especially as AI that makes up information starts to become more prevalent. > — [hatthew on hackernews](https://news.ycombinator.com/item?id=49027108) > I'm genuinely curious what you thought "AI restoration" was doing if the fact that it added details that it thought were reasonable was surprising to you > — [Bratmon on hackernews · 1 comments](https://news.ycombinator.com/item?id=49024890) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49021569) · 35 points · 36 comments ## Similar posts on daily.dev - [Your Old Photos Aren’t Lost Yet: What AI Photo Restoration Can Actually Fix in 2026](https://daily.dev/posts/your-old-photos-aren-t-lost-yet-what-ai-photo-restoration-can-actually-fix-in-2026-v3ii2lb9b) · Medium · 0 upvotes · 0 comments - [Teaching AI to See the World More Like Humans Do](https://daily.dev/posts/teaching-ai-to-see-the-world-more-like-humans-do-svbemqel2) · DeepMind · 0 upvotes · 0 comments - [How I brought old family photos to life with AI colorization](https://daily.dev/posts/how-i-brought-old-family-photos-to-life-with-ai-colorization-fozgyjmfk) · All Things Open · 0 upvotes · 0 comments --- Tags: [#chatgpt](https://daily.dev/tags/chatgpt), [#ai-image-generation](https://daily.dev/tags/ai-image-generation) [View this post on daily.dev](https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac) ```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":"What happens when the information runs out","url":"https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac"},"datePublished":"2026-07-23T16:38:04.933Z","dateModified":"2026-07-24T07:07:58.918Z","description":"An exploration of AI colorization applied to old family photographs reveals a fundamental limitation: when source information is missing, AI fills gaps by...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c5ab25a1f4ce8ff8b792f10503c5ec12?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c5ab25a1f4ce8ff8b792f10503c5ec12?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","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":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/what-happens-when-the-information-runs-out-gzgdzl6ac","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"chatgpt,ai-image-generation","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"What happens when the information runs out"}]} ```