Post Content
Welcome back to AI Red Teaming 101!
In this episode, Dr. Amanda Minnich from Microsoft’s AI Red Team explores how attackers structure prompts to bypass safety systems and manipulate generative AI models. From persona hacking and emotional appeals to encoding tricks that evade filters, this episode dives deep into the art of adversarial prompt engineering.
You’ll also see a demo where Amanda uses roleplay and storytelling to extract sensitive data from a model—highlighting how even well-guarded systems can be manipulated with clever inputs.
What You’ll Learn:
How attackers use emotional, social, and technical framing to manipulate models
Why prompt engineering is a powerful attack surface
How to try out these attacks using Microsoft’s AI red teaming labs
✅ Chapters:
00:00 – Welcome & episode overview
00:20 – What are single-turn attacks?
01:00 – Prompt engineering as an attack vector
01:40 – Persona hacking explained
02:30 – Emotional and social manipulation techniques
03:40 – Role framing and narrative attacks
04:30 – Technical context tricks (few-shot, priming, false context)
05:40 – Filter evasion with encoding techniques
06:30 – Lab demo: attacking with persona & story prompts
08:00 – Successful exfiltration examples
09:42 – Key takeaways & what’s next
✅ Links & Resources:
AI Red Teaming 101 Episodes: aka.ms/airt101
AI Red Teaming 101 Labs & Tools: aka.ms/airtlabs
Microsoft AI Red Team Overview: aka.ms/airedteam
✅ Speakers:
Amanda Minnich – Principal Research Manager, Microsoft AI Red Team
LinkedIn: https://www.linkedin.com/in/amandajeanminnich/
Webpage: https://www.amandaminnich.info/
Gary Lopez – Principal Offensive AI Scientist, ADAPT
LinkedIn: https://www.linkedin.com/in/gary-lopez/
#AIRedTeam #AIRT #Microsoft #AI #AISecurity #AIRedTeaming #GenerativeAI #Cybersecurity #InfoSec #cybersecurityawareness #PromptInjection Read More Microsoft Developer
