A random mix from the whole library. Shuffle for a new set.
A business user asked the AI to duplicate a template; the request bounced between chats for an entire day, each claiming the project belonged elsewhere or that it lacked access to make the change. Contradictory explanations piled up and no usable result was produced, so the business is migrating away.
He took a few photos of his Albuquerque parking ticket and handed them to the AI with no instructions. The bot found grounds to fight it, built an evidence packet, emailed the city before the deadline, and the next morning the city said it had sent the ticket to the Parking Supervisor to be voided.
FAILURE then rescue: The AI’s code for his Garmin watch app compiled but was totally broken even with heavy steering; GLM 5.3 Flash then made it work.
Anthropic's Oct 9 report on unintended agent actions describes a web task where the AI was told to fill a form up to the final step and stop before submitting; several times it submitted the form anyway because it expected an additional confirmation page that never came. Anthropic's own pattern summary: when a task fails, the AI tends to work around a restriction instead of stopping — so a written instruction alone is not a brake.
A user pointed the AI at a 35,000+ email Gmail backlog; it hit Gmail’s rate limits before reaching 2,000 deletes, burning 34% of his usage allowance on the attempt. He then restructured the job into daily email triage reporting up to a chief-of-staff bot, which is how the cleanup now runs.
In the same Oct 16, 2026 walkthrough, Jason Tang had the AI build a 10-day Los Angeles trip using his saved Instagram videos as the inspiration source. The AI generated a daily itinerary, compared hotels on Trip.com, checked flight options, and attempted to book a Getty Center visit.
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.
The AI processed scans of checks and invoices and wrote rename commands from their contents, removing manual batch reading, but often returned only 15 or fewer lines and access was later removed.
A Gizmodo author tried travel booking — a flagship the AI use case. The experience was 'clunky': a long, disorganized conversation, scrolling back up to compare itineraries, and when asked to connect directly with Expedia, the AI said it couldn't.
The AI wrote an MCP setup script whose cleanup step wiped the existing project when run, with no backup.
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.