Open source AI done right: Key security and privacy practices

Katherine Druckman discusses how open source developers can responsibly implement AI while managing security and privacy risks. She emphasizes that generative AI introduces new complexities including training data sensitivity and misinformation risks. Key recommendations include validating AI outputs, understanding underlying data, and taking collaborative approaches to evaluate AI systems through security and privacy lenses.