4 Oct 2025 Vladimir Buldyzhov
Protecting AI systems from advanced threats. Assessment, secure architecture, red-teaming for AI, and 24/7 monitoring
On September 25, 2025, H-X Technologies’ CEO Vladimir Buldyzhov took the stage in Kyiv at ANTI-FRAUD FORUM-2025 to present a hands-on framework for securing AI systems — from threat modeling and secure design to AI-specific pentests and production monitoring.
Key takeaways:
- Dual-use reality. AI boosts anti-fraud capabilities yet introduces fresh attack surfaces such as adversarial examples, data poisoning, prompt injection, and model inversion.
- Standards first. NIST AI RMF helps operationalize governance, mapping, measurement, and risk management for trustworthy AI.
- Defense in depth. Secure architecture, ML-aware DevSecOps, LLM audits, AI red-teaming, and continuous monitoring work best together.
Resources:
“AI security is a discipline, not a feature. We combine risk assessment, secure architecture, AI red-teaming, and continuous monitoring so organizations can make reliable, auditable decisions,” — Vladimir Buldyzhov, CISSP, CEO of H-X Technologies. |
Learn more about our AI Security service.
Book a free consultation.