AI and the New Era of Cybersecurity Threats with Mateo Rojas-Carulla
Today, we're exploring a topic that's becoming more important because it poses new security challenges: the rapid adoption of AI and the cybersecurity threats it brings. Our special guest is Mateo Rojas Carulla, an AI expert, chief scientist, and founder of Lakera. With an impressive background that spans Meta and Google and a journey from finance to AI, Mateo is a leading voice in the AI-first product revolution. We delve into how industries, even those heavily regulated like healthcare and finance, are increasingly integrating AI into their core strategies. Mateo will shed light on the unique security concerns that large language models (LLMs) introduce, including real-world examples of prompt injection attacks and the innovative defenses being developed, such as Black Arrow Guard. Additionally, we discuss the role of AI in testing AI, the fascinating AI security game 'Gandalf, and how these evolving dynamics necessitate a fundamental reevaluation of our security frameworks. Whether you're curious about AI's implications for productivity, the future of cybersecurity, or actionable advice for integrating AI-specific vulnerabilities into your security practices, this episode is packed with insights you won't want to miss.