How to Protect Your AI Model From Unwanted Conversations

People can write different prompts to an AI model and get these like undesirable responses from the model. There are a lot of other kind of like new security and vulnerabilities that we should be thinking about, he says. The ones that are closer to cybersecurity vulnerabilities have to do with what we call the software supply chain. In traditional cybersecurity, it's pretty common to look for known vulnerabilities inside open source code.