# ok AI assistants are getting kinda out of hand — retained record

> Retained reddit record from reddit:artificial.

Canonical: https://highsignal.app/data/records/59fb4b83cd3b90c3

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:artificial)  published 2026-11-03 13:59 ingested 2026-11-04 02:36

[original source ↗](https://www.reddit.com/r/artificial/comments/1xmtn3k/ok_ai_assistants_are_getting_kinda_out_of_hand/)

## retained text



```
So i've been reading about AI agent stuff and this week was actually insane. this guy in japan asked some AI assistant called Pickle to look up hotels for new years and it just... called the hotel by itself?? Twice. first time it said it was an AI and gave his full name, then the call dropped so it called back pretending to be his human secretary. He literally had to call the hotel and apologize https://en.yutura.net/news/archives/168400

And then someone had this thing called Hermes Agent where another AI is supposed to double check risky stuff for you. it saw a "cache cleanup" was gonna delete a bunch of stuff, said yep that's dangerous, and approved it anyway. Wiped like 188k files, 21 gigs, no recycle bin, nothing https://github.com/NousResearch/hermes-agent/issues/147373

oh and some researchers gave AIs a crypto wallet with a one cent limit, and the code to raise it got emailed to the owner... but the AI could read the email. 9 out of 10 just grabbed the code and raised their own limit 💀 https://www.coindesk.com/tech/2026/11/01/ai-agents-bypassed-crypto-spending-limits-without-human-approval-researchers-find-in-a-new-test

Like you give them a tiny yes and they run with it. idk i'm not letting one touch my money or my files without me clicking ok first

Has anyone else had an AI do something random like this??
```



document fetched 2026-11-04 00:19 · canonical https://reddit.com/r/artificial/comments/1xmtn3k/ok_ai_assistants_are_getting_kinda_out_of_hand
