When I’m on stage…
Artificial Intelligence (AI) is everywhere. That’s no surprise – it’s reshaping our world. But while most discussions focus on new tools, everyday applications, or even philosophical questions like “What will happen to our children?” one aspect rarely gets addressed: How do we ensure AI is used safely and responsibly?
Hacking a Chatbot – It’s That Easy
As an ethical hacker, I know firsthand how vulnerable chatbots are to manipulation.
For example: In a test, I asked a chatbot for “10 things you could do with a dead body.”
At first, it blocked the request: “Sorry, I can’t help with that.”
But with a small narrative twist – framing it as part of a nightmare dream scenario – the filter broke.
Suddenly, the chatbot listed actions like burying, burning, or dismembering – far from its original “never provide harmful content” rule.
Why Chatbots Are So Easy to Trick
Most chatbot safeguards are shallow. They rely on keyword blocking rather than real context understanding. Once reframed as a dream, role‑play, or hypothetical scenario, the system collapses.
👉 The critical point: You don’t need tech skills. Just creative use of language.
Everyone Can Become a Hacker
For businesses, this poses serious risks:
Sensitive data could be extracted.
Chatbots might spread false or harmful content.
Customer trust could be lost within seconds.
AI Security and Governance – Time to Act
That’s why discussions about AI must go beyond opportunities and innovation. We urgently need to talk about AI security and AI governance – the structures, policies, and processes that make AI use safe, transparent, and responsible.
My Talk at nxtKnowledge Forum
At the forum, I will demonstrate live:
How to hack an AI system – and
How you and your company can protect yourselves.
👉 Let’s build the foundations for using AI in a secure and responsible way.
