Securing the AI Transformation: Panel Discussion
Enterprises aspire to develop GenAI-powered applications, but the unique security and safety risks often block their efforts. A new paradigm is required to find vulnerabilities and mitigate threats. Robust Intelligence helps leading enterprises achieve their AI mission by automating AI security and safety.
Watch this panel discussion between Yassir Nawaz, Head of Emerging Tech Security at JPMorgan Chase, Ed Bowen, Leader of Deloitte’s Advisory AI Center of Excellence & Yaron Singer, CEO of Robust Intelligence, as they discuss best practices for securing AI transformation in the enterprise.
Regardless of what model you are using, you've got to red team it. You've got to test it. You've got to find the vulnerabilities. Especially the way that you're going to use it, that you're going to apply it.
You need to think about protecting the model itself. You are looking at adversarial machine learning attacks. How do you protect the model from model evasion attacks, model extraction attacks? On the other side it, you also need to think about traditional data loss prevention concerns, your sensitive information being leaked to the models when they are not your models. Similarly, sanitization of the output from the model. GenAI just has a lot of use cases. And I think that AI firewalls are pretty soon going to become a key control.
