#488: AI attacking AI is here (The New AI Red Team)

Big thank you to Cisco for sponsoring this video! Hackers are hacking AI models. Prompt injection attacks are happening all the time. AI's are hallucinating and giving incorrect information. The AI models you download could be made by hackers. Your users are posting confidential information like passwords and API keys into online AI models. Developers are leveraging AI systems in their applications without checking that the AI models are not open to prompt injections. Read more here: https://blogs.cisco.com/security/cisc... We need a way to protect AI systems. And Cisco have a solution. // DJ Sampath's SOCIALS // LinkedIn: / djsampath Twitter/X: / djsampath // David's SOCIAL // Discord: discord.com/invite/usKSyzb Twitter: www.twitter.com/davidbombal Instagram: www.instagram.com/davidbombal LinkedIn: www.linkedin.com/in/davidbombal Facebook: www.facebook.com/davidbombal.co TikTok: tiktok.com/@davidbombal // MY STUFF // https://www.amazon.com/shop/davidbombal // SPONSORS // Interested in sponsoring my videos? Reach out to my team here: sponsors@davidbombal.com // MENU // 0:00 - Coming up 0:49 - Securing A.I. 01:23 - The dangers of downloading open-source A.I. models 06:29 - Securing A.I. models 07:52 - The future of companies and A.I. 10:58 - Introducing Cisco AI Defense 13:33 - How to break an A.I. model and how to prevent it 16:08 - One-time protection 17:48 - Securing A.I. access 18:17 - What Cisco AI Defense provides 19:09 - Will Cisco AI Defense block attacks? 22:29 - The effects of Data Poisoning 24:38 - How will Cisco AI Defense be deployed 26:18 - When it will be available 26:30 - Conclusion Please note that links listed may be affiliate links and provide me with a small percentage/kickback should you use them to purchase any of the items listed or recommended. Thank you for supporting me and this channel! Disclaimer: This video is for educational purposes only.