How Attackers Trick AI: Lessons from Gandalf’s Creator
The AI Native Dev - from Copilot today to AI Native Software Development tomorrow
The AI Native Dev - from Copilot today to AI Native Software Development tomorrow
How Attackers Trick AI: Lessons from Gandalf’s Creator
Mar 18, 2025 Season 1 Episode 45
Tessl

🔒 How Secure is AI? Gandalf’s Creator Exposes the Risks 🔥

AI security is under attack, and hackers are finding new ways to manipulate AI systems. In this episode, Guy Podjarny sits down with Mateo Rojas-Carulla, co-founder of Lakera and creator of Gandalf, to break down the biggest threats facing AI today—from prompt injections and jailbreaks to data poisoning and agent manipulation.

What You’ll Learn:
- How attackers exploit AI vulnerabilities in real-world applications
- Why AI models struggle to separate instructions from external data
- How Gandalf’s 60M+ attack attempts revealed shocking insights
- What the Dynamic Security Utility Framework (DSEC) means for AI safety
- Why red teaming is critical for preventing AI disasters

Whether you’re a developer, security expert, or just curious about AI risks, this episode is packed with must-know insights on keeping AI safe in an evolving landscape.

💡 Can AI truly be secured? Or will attackers always find a way? Drop your thoughts in the comments! 👇

Watch the episode on YouTube: https://youtu.be/RKCvlJT_r4s

Join the AI Native Dev Community on Discord: https://tessl.co/4ghikjh

Ask us questions: podcast@tessl.io

Episode Artwork How Attackers Trick AI: Lessons from Gandalf’s Creator 54:35 Episode Artwork AI Gone Rogue? LLM Werewolf Showdown 54:43 Episode Artwork Monthly Roundup: AI Model Wars, GPT-4.5 vs. Sonnet 3.7, and the Future of AI Dev Tools 42:09 Episode Artwork The Future of Audio AI: Insights from Mati Staniszewski of ElevenLabs 1:02:52 Episode Artwork Building the Ultimate AI-Powered Development Environment with Farhath Razzaque 36:16 Episode Artwork DeepSeek R1: Ask Me Anything - Open Weights, MoE innovations, Model Distillation and more! 33:18 Episode Artwork Live Monthly: News on DeepSeek, Stargate, StackBlitz (Bolt.new) funding, prompting, and more 51:46 Episode Artwork RAG beats Fine-Tuning in learning your code base - but it doesn't make AI a better developer -Founder of Augment, Guy Gur-Ari explains 57:13 Episode Artwork Adapting your team culture for AI: DevOps Lessons Learned with Hannah Foxwell 42:21 Episode Artwork From AI Assistants to Agents: How Sourcegraph is Transforming Enterprise Development 45:17 Episode Artwork LIVE 2024 Year in Review: AI Innovations and Predictions with Guy Podjarny 51:20 Episode Artwork Mastering LLM Prompting in the Real World by Macey Baker 38:32 Episode Artwork Crossover episode with The Infra Pod - AI Native Development with Guy Podjarny 56:17 Episode Artwork The Evolution of v0 and Vercel's AI SDK, with Malte Ubl, Vercel CTO 25:43 Episode Artwork AI Evaluation and Testing: How to Know When Your Product Works (or Doesn’t) 49:58 Episode Artwork DevOps with AI: Identifying the impact zone, with Roxane Fischer 28:13 Episode Artwork Live Roundup: AI Myth busting in the real world and more with Matt Biilmann, Ben Galbraith, Patrick Debois and Simon Last 41:27 Episode Artwork Building Notion AI: Lessons Learned and Myths Busted with Simon Last, Notion Co-Founder and CTO 52:25 Episode Artwork Tessl Raises $125M to Build AI Native Development 36:02 Episode Artwork Beyond Coding assistants: Cursor as an API, Coding with gestures and more with Patrick Debois 36:47 Episode Artwork Does AI threaten the open web? Challenges and Opportunities with Netlify's CEO & Co-Founder, Matt Biillmann 55:30 Episode Artwork Live Roundup: Embracing AI in Development and Infrastructure, with Liran Tal, Amara Graham, Armon Dadgar and Patrick Debois 48:19 Episode Artwork Changing the Developer Documentation UX Workflow using AI with Amara Graham 32:42 Episode Artwork From DevOps to AI: Patrick Debois Shares Strategies for Successful AI Integration and Cultural Change 30:03 Episode Artwork Can AI Tools Be Trusted with Security-Critical Code? Real World AI Security Risks with Liran Tal 37:35