SN 1092: Restraint Abliteration

New episode of Security Now: Episode #1092, “Restraint Abliteration,” is now available.

Hosts: Steve Gibson and Leo Laporte

Topics covered:

  • :robot: A dangerous open-source AI proxy that could compromise users
  • :france: France’s under-15 social media ban hits constitutional challenges
  • :bug: AI-driven prompting uncovers a serious Zoom vulnerability
  • :chart_increasing: AI-based network defense companies see stock boosts after a major DDoS event
  • :brain: Deep dive into how AI chatbots evolved from simple prediction to today’s LLMs — and why safety, alignment, and restraint removal matter

#SecurityNow #Cybersecurity #TWiT

A PDF of the show notes should be accessible on this link:

Great show, i can easily imagine referring back to it in the future when it comes to AI methodologies.