MITRE ATLAS Explained: AI Security Threats + OpenAI vs. DeepSeek Case Study
from Simone/'s CyberSecurity Hosted by gtyu
Understanding AI Security with MITRE ATLAS + OpenAI vs. DeepSeek Case Study MITRE ATLAS is an attack matrix designed specifically to address security threats in artificial intelligence (AI) and machine learning (ML) systems. In this video, we break down: What is MITRE ATLAS? How it helps analyze AI-related cyber threats. How AI models can be exploited through API abuse, model distillation, and adversarial attacks. Real-world case study: OpenAI vs. DeepSeek – how OpenAI accused DeepSeek of shadow training by extracting AI model responses. MITRE ATLAS Mapping: We systematically map this incident using MITRE’s AI threat framework. Key takeaways: Why AI security is crucial and how companies can defend their AI models. Learn how attackers can misuse AI APIs and what security measures can prevent model exfiltration! Useful Links: Explore MITRE ATLAS here: https://atlas.mitre.org/matrices/ATLAS Check out the MITRE ATT&CK Navigator Tool here: https://mitre-atlas.github.io/atlas-navigator Let me know your thoughts in the comments! Should AI companies be worried about model exfiltration? Subscribe for more cybersecurity insights! #aisecurity #MITREATLAS #cybersecurity #machinelearning #artificialintelligence #openai #deepseek #aithreats #CyberThreats #apisecurity #techexplained #DataExfiltration #cybersecurityindia #aibharat
Embed this episode
No transcript yet.