Data Poisoning & Integrity - 20 de mayo de 2026 - TecnoWebinars.com

Data poisoning is a critical cyberattack that targets the training phase of machine learning (ML) and artificial intelligence (AI) systems, where malicious actors intentionally insert, modify, or delete data to corrupt the model's learning process. By compromising the integrity of this training data, attackers can induce biased, harmful, or incorrect outputs that may go unnoticed until the system is deployed. As AI reliance grows, data poisoning has shifted from a theoretical risk to a practical threat, with studies showing that poisoning as little as 0.001% to 3% of data can significantly degrade model performance In this session, we’ll discuss: - Core Concepts of Data Poisoning - Types of Data Poisoning Attacks - AI Integrity & Business Impact - Prevention & Mitigation Strategies