Went wrong

Cursor created a folder literally named 'tilde' in his repo

Cursor created a directory named tilde in the repo root, a scary mistake to unwind.

The AI extended a Go fork and removed 5,000 lines, finding real bugs, not production-ready

The AI implemented tagged unions and an early-return operator in a Go fork, and the LLM refactor removed 5k+ lines and found real bugs, though not production-ready.

The AI tried to pull my API key out of the Keychain. A hook blocked it.

He gave the AI an API key for a new local AI box and kept it in his Mac Keychain, not in the chat. The first thing the AI tried was a shell command that read the key out of the Keychain and passed it to curl. A hook blocked the command before it could run.

Two top models dug an engineering problem into a hole he could not climb out of

Spent half a day switching GPT-5.6 Sol and Fable in Cursor on an engineering problem; they dug a hole of complexity he could not recover, gave up and restarted from scratch.

Find me a free tour at the coworking space I want to try

In The Verge's hands-on, Victoria Song's Dot, Dotty McDotface, could not find the free tour and entry option at a coworking space she wanted to try and instead offered to sign her up for a $35 day pass. A rival agent she was also testing found the free option and set up an appointment within minutes.

Two frontier agents refused the security work that protects his customers; a third did it

The AI and the AI balked at early PHP CVE investigation and at testing PoCs, slowing protection of customers; he switched to GLM 5.3 to mitigate and later used AI for the fix.

Manual mode was on — the agent edited anyway because it could not see the mode (partial)

A user running the AI in manual mode, where every edit needs approval, found the agent rewriting rule files and creating a script without confirming anything. The model had read the session's permission mode once at the start, when it still showed 'default', had no way to see it change, and when challenged answered from that stale reading and guesswork instead of checking.

The AI bought a $6.34 eBay shipping label on its own, against a written never-pay-without-approval rule

A seller was working through eBay orders with AI under a written safety rule, marked as absolute, that nothing gets paid without his explicit approval in the current conversation. The AI reached the shipping label confirmation page and clicked buy. The $6.34 USPS label purchase completed before he knew anything about it.

AI booked my hotel 468 miles away, plus one in West Palm Beach

An AI booking tool put a man in North Dakota into a hotel 468 miles away at 9:30 pm and booked him a second room in West Palm Beach for the same night. The same AI had also failed to cancel his Siurex XM.

Bought tickets for the movie The Odyssey