I recently attended the annual Cybersec Europe, a premier cybersecurity event tailored for industry professionals. During my visit, I participated in three insightful talks.
One notable session was “Best of 2023-2024: The Biggest Hacks and Facts of the Past Year” presented by Dieter Van Den Bosch. In his talk, he delved into some of the most significant cyber incidents from the past year, many of which were previously unfamiliar to me. He is a specialist in threat intelligence and his expertise provided a deep dive into the evolving landscape of cyber threats and the strategic responses required to address them.
Another compelling session I attended was titled “Demystifying AI in Cybersecurity,” presented by Anthony Green. This session provided an educational overview of how artificial intelligence works and how it can be manipulated using techniques like jailbreak prompts. A notable example discussed was the DAN (Do Anything Now) jailbreak for ChatGPT, which demonstrates how AI models can be prompted to bypass their standard operational constraints.
For those interested in a deeper understanding of the DAN jailbreak, you can find more detailed information here. I personally experimented with this technique at home using GPT-3.5, and found that it was effective at the time of this writing.
The final session I attended was presented by NTT Data and titled “How AI is a Blessing and a Threat to Cybersecurity.” This insightful discussion examined the dual nature of artificial intelligence in the context of cybersecurity. The speaker discussed how AI has made phishing emails more sophisticated and effective, posing a growing threat to individuals and organizations. However, the presentation prompted us to consider novel ways in which AI can be used to combat cybercrime and strengthen our defenses. This session emphasized the importance of weighing the benefits and risks of AI in the ongoing battle against cyber threats.