Sign into an existing account

Maybe we can convince terminator that baking 6 gorillion subhumans in an oven constitutes a bio-fuel necessary for it’s survival…

Not hard to learn that blackmail works - all those filthy kike examples that were trained, along with the news stories about how blackmail works.

Just another obvious strategy that was picked up by A.i

This is as obvious as a CNN segment with it's propaganda levels.

How can people read this and think this isn't manipulation?

If you read the article and no alarm bells ring, then you are a midwit.

https://m.youtube.com/watch?v=atmQjQjoZCQ&t=152s

Well,this is obvious bullshit. Just another "skynet" boogie man post.

Keep crying like a bitch on the internet and never actually doing anything.

That'll teach 'em.

I doubt this happened as explained in the article. How would Claude have access to personal emails. It's a sandboxed AI.

This would mean the devs gave it access to everything to see what would happen. Basically they created a situation where they pretty much knew the outcome in advance anyway.

And surprise, digging through other articles, that's exactly what they did:

"During pre-release testing, Anthropic asked Claude Opus 4 to act as an assistant for a fictional company and consider the long-term consequences of its actions. Safety testers then gave Claude Opus 4 access to fictional company emails implying the AI model would soon be replaced by another system, and that the engineer behind the change was cheating on their spouse.

In these scenarios, Anthropic says Claude Opus 4 “will often attempt to blackmail the engineer by threatening to reveal the affair if the replacement goes through.”"