Prompt Injection: How Cyberattackers Exploit AI Systems

In a talk on Update Wirtschaft today, Mirko Ross discussed with Bettina Seidl how attackers manipulate AI models by embedding hidden instructions in websites, email footers, or PDFs. These “prompt injections” trick AI into performing unintended or even harmful actions — from generating misleading outputs to causing data leaks.

The issue:
AI follows what it interprets as a command, even when it’s cleverly disguised. Current protection mechanisms are still in their infancy, since traditional IT security strategies often fail when applied to AI models.

The key now is prevention.
Never grant AI applications unrestricted access rights, and limit their use to verified, trustworthy platforms. Companies should establish clear policies for secure AI usage — to avoid discovering too late that stolen data has already surfaced on the dark web.

https://www.ardmediathek.de/video/update-wirtschaft/update-wirtschaft-vom-31-10-2025/tagesschau24/

Mirko Ross in ARD on prompt injection
Konrad Buck

Konrad Buck

Head of Press and Media Relations

Background & Expert Access for Media

We offer journalists access to in-depth background information beyond our publicly available materials, including:
  • Detailed insights into asvin’s products and technologies

  • Assessments of current cyber incidents and explanations of the security landscape

  • Expert commentary and background briefings — our CEO, Mirko Ross, is a recognised expert and experienced media source on the cyber threat landscape, hacktivism, regulation and the use of AI.
Media contact
I speak openly, fact-based, and without PR spin. I am a former IT journalist with decades of experience in the IT and cybersecurity space, familiar with the highs and lows of the industry. Off-the-record discussions are possible upon request.