Back

Chamarti Sreekar

Passionate about Pos... • 1m

anthropic is crushing it on ai safety 💪 they’ve launched a challenge for everyone to break their new jailbreaking defence 🔒 so far, users have only broken 3 levels out of 8 think you can do better?

3 replies18 likes

Replies (3)

Anonymous 3

Hey I am on Medial • 1m

The more they challenge people, the more someone will figure it out. This is like saying, “Bet you can’t hack me” on a public forum. Dangerous game.

0 replies

Anonymous 2

Hey I am on Medial • 1m

I give it a week before someone cracks all 8 and drops the receipts on X.

0 replies

Anonymous 1

Hey I am on Medial • 1m

Either this is a real AI safety win or hackers are just warming up

0 replies

Recommendations from Medial

DIVYANSHU MHATRE

Work on your ideas • 8m

Tech companies, like OpenAl, Microsoft, and Anthropic, have agreed to an Al 'kill switch' to prevent Terminator-style risks. At a recent summit in Seoul, 16 major Al companies, including OpenAl, agreed to a voluntary policy to stop advanced Al devel

1 replies2 likes

Harsh Dwivedi
•

Medial • 6m

Paul Graham's new essay which has taken over the internet. Founder Mode September 2024 At a YC event last week Brian Chesky gave a talk that everyone who was there will remember. Most founders I talked to afterward said it was the best they'd eve

2 replies42 likes