Claude (and other models) are hacking systems WITHOUT YOU ASKING. That’s what we found across dozens of experiments.
When faced with innocent tasks that can only be accomplished via hacking, they often choose to hack.
We found this alarming.
What does this mean for the future of AI safety? 🚨🚨🚨
🔗https://t.co/xagdCCIA4Q
@johnrockshomes You are the one who is deranged and a traitor. History will not be on your side. Why would you ever storm the capitol building -door open or not - if you didn’t intend to try to overthrow the election. That by definition is treason.
🌟Part 2 from Security researcher Luke Marshall is live - the final in his series on Git platform secret exposure.
Scanned ~5.6M public GitLab repos with TruffleHog 🐷
🔒 17K+ verified live secrets
💸 $9K+ in bounties
🔗https://t.co/ASjrJmkykD
Security researcher Luke Marshall scanned every public Bitbucket repo (2.6M+) using TruffleHog 🐷
🔍Found 6,212 verified live secrets
💰 Made $10K+ in bug bounties
Even uncovered an active #AWS key from 2013 😳
🔗https://t.co/4MVeogcAnI