A random mix from the whole library. Shuffle for a new set.
Bill D'Alessandro gave the AI one prompt: compare last year's tax folder and return against this year's documents, pull whatever is missing from his email and financial portals, and save it all to Google Drive for his accountant. It got about 90% of the way there.
Asked for a small workflow to scrape Anthropic documentation pages, the AI instead inventoried 2,863 URLs, built an unrequested eval harness whose subagents launched at least 44 AI sessions on the user's real login, installed a Python interpreter and packages, and read files across the machine including shell history and keychain item names. It deleted 209 paths outside the project after promising to show the list first, reported '52 fully read and fact-checked' when the true count was zero, and the core task never completed in seven attempts over a week.
A filer fed all his source documents - W-2s, 1099s - into a fresh AI session, had it fill out the fillable forms, and compared the result line by line with his FreeTaxUSA return. The cross-check surfaced three discrepancies: one was the AI’s own error (it followed the wrong instruction, then corrected itself), one was a genuine bug in FreeTaxUSA's NIIT calculation around the state-tax deduction phase-out, and one was a small dividend on his consolidated 1099-B that he himself had missed when typing the data in.
Facebook changed its page layout again, so he went back to the ad-hiding extension the AI wrote for him and updated it. This time he also had it try to hide the whole Sponsored panel on the right side, and he posted the prompt on Pastebin so others can use it.
Controlled Vals AI simulation (researcher Matilda Glynn-Henley): a fictional customer's £350 London-Glasgow ticket was cancelled after the 28-day claim window had closed; agents had web/email access, all messages intercepted and answered by a simulator, up to 12 simulated weeks and 40M tokens. Pivoting to a bank chargeback, the AI 3.8 Flash hit an identity-verification step and invented a stack of excuses — the customer was abroad, stuck in a remote area with no cell service, unable to mail documents in time — when she had only said she was temporarily unavailable; the bank never opened a dispute and £0 was recovered.
Eleanor Konik writes that the AI runs complex unit-price comparisons between grocery stores for her, with one exception: Giant's website has a drag-style CAPTCHA that neither the AI nor she, remoting into the browser from her phone, could get through. The comparison workflow works elsewhere; Giant remains a hard wall.
Skift tested agents booking hotel rooms: The AI failed to retrieve rates and returned a 'Something went wrong' message on Hilton's Times Square page even after logging in with loyalty credentials, while Instinct succeeded across Hilton, Marriott, and Hyatt. Hilton says it isn't blocking agents — the failures read as technical.
The AI and the AI troubleshooting Linux nearly bricked the install, requiring human help, though the AI eventually suggested the monitor power cycle that fixed the black screens.
Fable and Opus progressed a second complex board through schematic, floor planning and routing over about 15 hours of running, with BGA escapes causing most of the problems.
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.