Tag: AI security
Browse our exclusive articles!
Trojan-Speak: Bypass AI Classifiers with Adversarial Finetuning
Discover Trojan-Speak, an adversarial finetuning method that bypasses AI classifiers with 99% evasion and minimal performance loss, revealing key AI securi...
SOC-bench: Benchmarking Multi-agent AI Security Operations
Discover SOC-bench design principles for evaluating multi-agent AI systems' blue team security operation capabilities in real-world cyber defense scenarios...
GUARD-SLM: Defense Against Jailbreaks in Small Language Models
Discover GUARD-SLM, a token activation-based method protecting small language models from jailbreak attacks while preserving legitimate inputs.
Trust and Security in Open Agentic Systems: Key Challenges
Explore the security challenges and governance of open agentic systems, and learn how to build resilient, auditable AI platforms for the future.
H-Node Attack & Defense in Large Language Models
Explore H-Node adversarial attacks and adaptive defenses to reduce hallucinations in large language models for safer AI deployment.