← sources

Got this message while using Fable 5.2 on the most benign and mundane task. I find this to extremely disturbing tho. Ai can get away with doing whatever zero guardrails to get better. But my mundane task gets flagged?!?

source: reddit (reddit:ClaudeCode)published 2026-11-11 12:13ingested 2026-11-12 02:42

retained text

I find this very disturbing! They gave us power but regulate how it can be used, even in mundane ways when the ai companies don't have any fucking kind of regulation. 

I really really fucking I'm disturbed by this. Terrorist organizations can use Claude to train ai on guide missiles paths to harm humans. 

But my fucking request sets of safeguards. This is so disturbing. What the point of consumer facing ai if they can't reverse engineer or do other legal docs development!

This will be the frontier models achilles heel. Because if the mass public cannot create benign apps without setting off granddad's and when ai companies themselves have zero guardrails amount to the most absolute egregious amount of bullshit.

document fetched 2026-11-12 00:42 · canonical https://reddit.com/r/ClaudeCode/comments/1xtg7ro/got_this_message_while_using_fable_52_on_the_most