New AI Security Papers 🔥 and my TEDx talk on Hackbots

Hey everyone!

It’s been a busy time for my family and work, but I’ve got some really great stuff to share today! 😊 

1. There were two great AI Security papers that dropped recently. The first is on Jailbreaking Top LLMs. Look at the success rate (admittedly it was judged by GPT-4, so it’s not likely to be 100% accurate, but it’s still very impressive):
   

   

2. And the second is LLM Agents can Autonomously Exploit One-day Vulnerabilities. Neither of these are surprising to me, as I believe AI Agents Will Outhack Humans eventually, but still great to see official research in this area.

3. I was on Nahamsec’s youtube channel showing how I use AI inside of the text editor, vim. It’s a fun video!

4. I did a TEDx talk on The Rise of Hackbots at my alma mater, the University of Kentucky. This should link to the right timestamp, but if not just go to 2 hours and 53 minutes. The video quality isn’t great, but the content is! 😎 It’s only 10 min. Check it out here:

Thanks for being on the email list! 😊 If you like this content, I’d love if you invited someone to join it or to follow me.