Transcript
Lastly, let's close the loop on this - is this concept of generative AI, right? Any number of users may send content to, uh, you know, an AI engine, right? This may be source code, this may be, uh, you know, confidential content, uh, that you're trying to do, let's say you're trying to create a press release and you, uh, for an acquisition, and you send that to ChatGPT. Um, or you may send some, some pipeline numbers, say, hey, I analyzed these pipeline numbers. Users, again, are just trying to be productive, but they don't realize that that goes into, uh, the ChatGPT cloud on the right. And then it becomes very much aware that this type of, uh, information, if I ask, tell me about Acme, uh, you can learn a lot, right? So, what Zscaler can do is block all of those things and say, okay, I'm going to allow you to use ChatGPT, but I'm going to use DLP inspection to decide if that content is sensitive or not, or cloud app control or URL filtering, and then allow you to filter what goes all the way to the right ChatGPT. Um and then the end result is you can enable users to use that, but prevent sensitive data going into the big brain in the sky, which then allows you to control your data so that, um, you know, the public and other people searching for information only find what's publicly available, not not specific information.
