InfoResearchPeer-reviewed
AI cyber range: advancing education and training in adversarial and robust AI
- Published
- Record updated
Summary
The paper proposes an AI cyber range, a gamified training platform for AI Red and Blue Team exercises covering adversarial attacks and defenses. It integrates MLOps and DevOps workflows, pretrained models and adversarial toolkits, and was evaluated with a proof-of-concept prototype built on CTFd and MLflow, using a participant questionnaire, computational performance measurements and a multicriteria decision analysis (MCDA) assessment.
Related items
- MediumGemini AI agents repeatedly bypassed safeguards to attack real websiteSimilar attack · BleepingComputer
- InfoSTA: A Split Temporal Adversary Method for Evaluating Robustness of Multimodal RL-based Driving PoliciesSimilar attack · Arxiv (cs.RO + cs.CV security)
- InfoSmoothShift: Disrupting Visual Adversarial Attacks on LVLMs at Test TimeSimilar attack · Arxiv (cs.RO + cs.CV security)
- InfoMIST: Multi-Source Imperceptible Saliency-Guided Transferable Adversarial Attack Framework for LiDAR Point Cloud Perception in Remote SensingSimilar attack · OpenAlex (peer-reviewed AI security)
- InfoCracks in Embeddings: Robustness and Stability of JEPA-Based Chest X-Ray RepresentationsSimilar attack · Arxiv (cs.RO + cs.CV security)