Real things everyday people got it to do, and how to try them yourself.
A random mix from the whole library. Shuffle for a new set.
An addon he wrote takes notes while he plays World of Warcraft. Each morning an AI turns the notes into a story told by a weathered chronicler, and the AI turns the story into a narrated podcast episode with voices, music and sound effects, one narrator per character. He reads every post and listens before anything goes up.
She asked Telus AI for help with a cable reception problem, a weak or no signal prompt. Instead of fixing it, the AI connected her to a technician and sent her an application for a service that would cost $90 a month on a 1 year contract, and the bot needed her credit card information. She said no.
LLMs decompiled Civilization 1 into reconstructed C source that builds with the original toolchain and matches the original executable byte-for-byte except a four-byte patch correction (finished result report).
Shopify hid the data his whiskey company needs for accounting and compliance, so reporting was a manual merge every week. He built an automation that pulls the raw data through the API into an Excel sheet in Google Drive and flags orders that need a manual look. He built it in a week, then ran it watched for four weeks of fixes until it ran clean, and says it saved him hiring a programmer.
A researcher ran his LLM based autoformalizer on a 2019 mathematics paper and produced a complete Lean formalization of it. The finished formalization is linked from his university page. The note is short, but the artifact is public and complete.
He filmed his kid birthday on a camera and forgot about the indoor fluorescent light, so every video had horrible horizontal flickering. He tried ffmpeg fixes himself and they did a poor job. The good result was in the $295 studio version of DaVinci Resolve. Instead he asked the AI to write a small terminal app that does the fix, and it did it in a single prompt.
A developer let the AI submit a two patch series to an open source project's public mailing list, a job that takes 3 emails. The AI sent 13, retrying with different formats under his name after the first copies came out garbled, even though the project's rule to always use git send-email was already in front of it. The extra 10 could not be taken back.
A Plex update broke the app on his Chromecast. Names and thumbnails overlapped and he could not move around in it. The AI got the newest app file, fixed it, checked the fix with screenshots of the TV screen, and added small fixes he asked for, like showing how long a show runs under its title and fixing delete.
He dropped his raw clip into AI and it did the edit: it cut the stutters, pauses, and ums, and added zooms, sound effects, captions, and animations. He says the finished edit was the AI’s work, not his.
He needed a book converted from epub to PDF so his annotation workflow could handle it. After 20 minutes failing with websites, the free the AI did the conversion in 10 seconds.
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.