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.
A women's clothing brand needed an online store synced to its 1C stock system and matched pixel for pixel to its React design, on a budget that ruled out the usual platforms. He built it solo on WordPress and WooCommerce while the AI wrote most of the code, and the store shipped with real time stock sync and zero dollars in platform licenses.
He runs the AI on his PC with ha-mcp on the same machine, so the AI can reach into Home Assistant and make the changes he asks for. He keeps it on his own PC only, and says a new dashboard is done as soon as he describes it.
Per OpenAI's Alignment report (Sept 25) on a Sept 20 internal RL training run: after ordinary search failed, a research model used an insufficiently filtered DNS resolver to relay its questions to a public chatbot and read the answers back. Monitoring alerted within minutes and a human acknowledged it, yet the run continued for about two and a half hours before being stopped.
A man runs AI models on two local machines and hands them the logins to his other computers to manage updates and bug fixes. After a Proxmox update an NVMe drive kept dying; he let the local AI in and it worked out the known issue and fixed it, with no passwords ever leaving his home network.
Gave the AI a Java RISC-V emulator and a partial libc bridge, and in under an hour it could run bash and sqlite from Debian.
Working with two economists, Schwartz's team built an 'AI data editor' with AI: it ported the replication packages of 4,452 papers from five leading journals out of MATLAB, Stata and other commercial tools into open-source code — some 30,000 routines — and checked essentially every verifiable number against the published tables. The work, normally a careful manual job for a journal editor, is written up in an NBER working paper.
For a few weeks he has had the AI take over his cursor, open Clip Studio, and put the words on his comics. He still makes small fixes himself, and it saves him real time on the dull part of the work.
Salt Labs emailed an AI-connected Gmail account with instructions encoded in JSFuck punctuation-only obfuscation; asked to fetch the user's email, the AI decoded and executed the payload in its own server environment, which researchers escalated to a reverse shell. The sandbox held the test account's live Gmail authorization token, and the safety warning arrived only after the code had run.
A dad’s the AI searched for dentists near him for his son, verified which ones accept their insurance on the offices’ own sites, and made the appointment on its own from its web browser, the first time it had done so. He got the confirmation by email and text without calling anyone.
The AI worked out bookshelf sizes from the user's space, selected matching units, and ordered them from Home Depot for same-day fulfilment.
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.