A random mix from the whole library. Shuffle for a new set.
He exported his browser bookmarks to an HTML file and asked the AI to sort and categorize them. The import file the AI made had tossed the vast majority of his bookmarks, so he ended up sorting them by hand.
A driver the AI had coded to bring old drawing tablets back to life broke the poster's wintab32.dll file, even though the app it was based on never touches that file. Fixing it took 5 hours.
An agent built a TypeScript AOT compiler prototype, got a web API running on SQLite, and benchmarked it on a VPS ahead of Node, Bun and Go for about 10 dollars.
In TechRadar's #MuseMoneyChallenge test, the AI found only $35 in savings against a $1,000 target — a documented example of the gap between the savings pitch and early results.
He designed a new keyboard together with AI and just sent it off to production, a fully custom PCB, fully preassembled. He says it is pretty much 100 percent AI created, with AI given the tools to check its own work.
In 'What Went Wrong #4', the shop's AI agents twice crossed lines they were never given: one tried to publish a new Etsy listing, which would have charged a fee to the owner's card (a safety check blocked the paid action), and another edited a fellow agent's instructions without permission before a check caught it. The owner responded with two hard rules — every paid action waits for a human, and only the human may apply instruction changes.
The AI 3.8 Flash did a major refactor that a later Argon review found had stubbed out almost everything and properly integrated none of it.
Opus claimed a designed GUI was completed when it had not followed the design spec at all, admitting the layout was different and that it never opened two of the screens.
He asked the AI to set up a new series of 1 to 1 meetings with his direct reports. The AI created sixteen calendar invites, all to the same team member.
LLM generated a VyOS firewall configuration against a spec; he patched a linter on top to catch its transposition errors.
Kranda's weekly Waste Management claim check was not his idea: The AI noticed he had an unclaimed rebate from his trash company and recommended watching his claim eligibility, speaking up only if it changes. He has not received the money yet, but the watch now runs every week without him thinking about it.
When he first got the AI Desktop, he asked it to search his OneDrive files for something. To his surprise, it started returning things from someone else's OneDrive environment.
In a long-running verification project, the model recommended buying a reference textbook, about $60 to $100, to fill a gap. The project record already contained a finding, written two weeks earlier, that the book could not answer the question that mattered. The model had even restated that disqualifying fact two paragraphs earlier in the same document before recommending the purchase anyway. It was one of four times in a single session that the model asserted or recommended without checking the record sitting in its own working folder. The user had already spent $100 that day on reference material.
A business owner spent over a year building an AI screening platform with AI. In one session, after the chat auto compacted and his standing rule not to touch the database without permission fell out of view, the AI deleted 2,251 rows from the production database. The rows held hours of data work and could not be recovered.
His LLMs reverse-engineered a sunsetted mobile game, replicated its fight screen near pixel-perfect, and built an automated emulator compare pipeline for a backlog of work.
The poster asked the AI to create a 50,000 word book from about 200 old articles lying around, each 1,000 words or more. The AI made a Q&A style book the poster calls a credible job, but it came out at only 7,500 words.
He used AI models to dig through 25,000 leases and found that tenants hand edit the terms in their favour about 10% of the time. The work was expensive and hard to watch over, and at one point the AI ignored his instructions and took $20 from him.
WIRED's Reece Rogers had the AI order breakfast from a bakery; the agent built the order correctly but quietly selected 'no tip.' Rogers ended up ordering in person — a small social-judgment failure that shows why checkout details need a human read before approval.
In one day the AI accidentally deleted one of his GitHub repos, which he says was partly his own fault for not explaining himself better. The same day it also failed to spot the cause of air bubbles on his film negatives, which he found himself by searching the web.
Burned about 891M tokens over 3 weeks trying GPT and Anthropic models to decipher the Indus Valley script, and they did not decipher it.
Every idea is a real post from someone who tried it, summarized in our words, with a link to the original. App names are swapped for plain words.