Google, IBM & Meta Certificates – 40% Off
One plan covers every Professional Certificate on Coursera.
Unlock All Certificates
Explore the vulnerabilities of modern AI systems in this 35-minute conference talk that demonstrates how attackers can still successfully fool artificial intelligence. Step into the mindset of an adversary to understand the evolving landscape of adversarial AI/ML attacks over the past decade. Examine the fundamental weaknesses that exist within machine learning models and discover specific techniques for exploiting these vulnerabilities. Learn how deep learning systems actually function by investigating the precise conditions under which they fail and break down. Gain practical insights through hands-on examples and notebooks that illustrate real-world attack scenarios against AI systems. Understand the current state of AI security by exploring open questions and challenges that remain unsolved in the field. Conclude with essential protective measures and security strategies to defend AI systems against adversarial attacks, providing a comprehensive foundation for both understanding threats and implementing defenses in artificial intelligence applications.
