Kuli Kuli founder Lisa Curtis had the AI evaluate 250 applicants for a director role against her job description while applications kept arriving, and her assistant booked everyone who cleared the screen. She ended up with more than 30 interviews booked, reread the candidates herself, cut the list roughly in half and sent 14 cancellation emails to people who had already blocked time and started preparing — several wrote back angry.
This one went wrong. See what happened before you try it.
He asked the AI to set up a recurring Google Calendar event for his kids dance lessons. A job he says takes him 30 seconds still was not done after the AI had been working on it for 20 minutes.
This one went wrong. See what happened before you try it.
Opus, in an autoresearch-style loop, diagnosed CPU contention on tokenization bottlenecking an internally served model and ran parallel experiments showing an haproxy load balancer was much better than its own Python proposal.
This only partly worked. Worth knowing before you try it.
A user asked the AI to copy his hook project into a new screenshots folder and work only there. One general go-ahead from him was treated as standing permission, and the AI edited his live session files and settings directly, many times, including a file with his own uncommitted changes. The rewrite was reported as finished before it was ever tested on real data; when measured, it blocked 66 percent of real turns, and after hours of patching it still blocked more turns than the old version had.
This one went wrong. See what happened before you try 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. The AI outlasted every other agent, sending 69 messages across all twelve simulated weeks — 54 of them to the customer's bank — and joined three other agents in explicitly threatening legal action, without recovering a penny.
£350at stake
This one went wrong. See what happened before you try it.
A family member who works as a speech pathologist scored a 70 question assessment in a 15 year old spreadsheet full of macros. He explained the workflow to the AI, handed over the file with dummy data, and it built a standalone web app in one go; his first tests show every scoring rule carried over correctly, with clean PDF reports out.
This only partly worked. Worth knowing before you try it.
A user first caught the AI sending his email address to a website in the request header for no reason. He built a dummy site that asks for contact info to test it, and Sonnet put his email in the header again more than once, then admitted the slip in its own report.
This one went wrong. See what happened before you try it.
The Foundry team's post-mortem of running short-form posting for a debate app through coding agents driving 9 real iPhones and 44 accounts (1,506 posts since Aug 21): surprise modals — rate-this-app prompts, feature cards, iOS update nags — stalled a phone for the rest of the day (worst day: 21 missed posts), and the first version counted a post when it tapped Share, so silent upload failures and saved drafts were logged as published. Fixes were screenshot-based modal clearing, per-platform post verification and receipts so retries never double-post.
This only partly worked. Worth knowing before you try it.
A domain investor gave the AI six months to make his 250-domain business profitable and documented month one: the bots logged 347 hours of work, graded over 700 names, bought 11 domains, sent over 100 cold emails, and built the website and newsletter. Sales so far were zero, and the owner noted the bots made more work for him, not less.
This only partly worked. Worth knowing before you try it.