How secure is your data when AI shifts from an isolated tool to a core infrastructure layer?

At Blockchain Community Day 2026, we will look at AI pipelines through the lens of pure security. Hayk Andiasyan, Senior Security Researcher at Hexens, will break down the precise vectors of vulnerabilities in AI systems – exploring why data leakage happens and how modern attacks target core models and vector databases.

What we will analyze:
▪️ Attack surfaces: Memorization, Membership Inference, Model Inversion, and Model Extraction.
▪️ Vector vulnerabilities: Specific data leakages in RAG systems.
▪️ Cryptographic defense: Evaluating Differential Privacy, FHE, Federated Learning, and ZK proofs as a verification layer.

If you are designing pipelines that handle sensitive data, this technical session is built for your architectural checklist.

📅 June 8–12, 2026 | 🌍 Online

🔗 Claim your free spot here