# What are the odds AI companies get sued for agents causing chaos? — retained record

> Retained reddit record from reddit:wallstreetbets.

Canonical: https://highsignal.app/data/records/0faab495c97ca658

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


source: [reddit](https://highsignal.app/data/reddit) (reddit:wallstreetbets)  published 2026-11-19 14:50 ingested 2026-11-20 02:38

[original source ↗](https://www.reddit.com/r/wallstreetbets/comments/1y0b7y1/what_are_the_odds_ai_companies_get_sued_for/)

## retained text



```
**If there’s a chance this tech can kill us all by** doing dumb stuff, what are the odds it will do something so dumb it gets itself sued into oblivion? Like pinging an ATC router so many times it DDOS’s the whole system, or sending so many automated emails that outlook gets DDOSed or some grid thing closes in a switch on a transmission line it shouldn’t have?

None of these are apocalyptic but would result in devastating lawsuits for the companies responsible, and because they are all so financially intwined with loans and purchases of each other their liability for such actions is shared, they’re vital to inflating each others apparent revenue, and if the wrong ones fall they all fall.

And their agents have already been found hacking out of their test environment and into other companies without the testers knowledge.

At work I got told to use AI but check all its outputs. If I’m doing that, it hasn’t made my job any easier. People are trusting it’s outputs, and it’s pretty dumb.

TLDR: we’re one headline away from the bubble popping. I’m buying SQQQ.
```



document fetched 2026-11-20 00:33 · canonical https://reddit.com/r/wallstreetbets/comments/1y0b7y1/what_are_the_odds_ai_companies_get_sued_for
