『The Fight for Secure AI: Countering Adversarial Tactics [4]:』のカバーアート

The Fight for Secure AI: Countering Adversarial Tactics [4]:

The Fight for Secure AI: Countering Adversarial Tactics [4]:

無料で聴く

ポッドキャストの詳細を見る

このコンテンツについて

AI systems are vulnerable in ways traditional software isn’t. In this episode, we enter the cybersecurity battlefield of AI, where adversaries exploit weaknesses in machine learning models. We’ll explore adversarial examples subtle manipulations that trick AI into making mistakes and model poisoning, where training data is corrupted to sabotage performance. We’ll also discuss AI red teaming, the practice of ethically hacking AI to uncover vulnerabilities before malicious actors do. From tricking Tesla’s Autopilot with stickers to the downfall of Microsoft’s Tay chatbot, you’ll see why defending AI requires new strategies that go beyond firewalls and encryption.#aiethics #aigovernance #TRISM #AISafety #aitrust #aisecurity #airisks #responsibleai #Cybersecurity #DigitalTransformation #AIEthics #airegulations #aiinclassroom #TechSecurity
まだレビューはありません