What if I told you that you can trick an LLM into revealing secrets, making bad choices or even acting against its own rules? AI may seem like a black box, but when you start testing it like any other system, surprising weaknesses start to appear.

How easily can an attacker manipulate an LLM? What happens when sensitive data leaks? Can a chatbot be turned into a security risk? I’ll answer these questions while using my own cartoons to illustrate key risks in a fun and easy-to-understand way.

The good news? You don’t need to reinvent the wheel to test AI. Strong system thinking, traditional testing techniques, and a critical mindset are already powerful tools for uncovering vulnerabilities.

 Main take aways: 

• Haters: Learn how to delegate (your hate) to people who will not flip tables 
• Terrified: Your fear is justified. Code is scary stuff!
• Skeptics: Get over yourself. Learn how to code. 
• Enthusiasts: Get over your tooling fetish (and spread your enthusiasm!)
• Capable: You’re ‘special’, please spread your wisdom. The world needs you.

Maryia is a Quality Engineering Leader with a focus on backend testing and embedded systems. She leads testing initiatives, drives education programs and actively contributes to the Swedish testing community. As a mentor, speaker and content creator on LinkedIn, she helps new test professionals build their skills and confidence.

She believes that great testing is about both technical excellence and knowing when to step back, think critically and enjoy the process.