Many of you know I’ve been working with Azure Open AI and the ChatGPT deployment models a lot recently in an effort to understand the security implications of running them in an organization. We all know that when C-levels and users see a shiny new thing they want it RIGHT NOW without doing the due diligence to ask the appropriate security implication questions. So, I’m spending much of my time generating lots of data to determine exactly what can and can’t be monitored and how our security mechanisms apply.
In doing this, I’ve had to also had step up to create some small apps here and there so I can generate the data I need.
One recent thing that I created, I thought I’d share with all of you to help drive your creativity around the things that are possible with Azure Open AI ChatGPT, is simple command-line Chatbot. The Chatbot uses a couple external files - one for storing configuration data (API info) and one that stores the question that will be posed to ChatGPT.
You can find all the necessary code and collateral here: https://github.com/rod-trent/OpenAISecurity/tree/main/Code/Command%20Line%20Chatbot
As noted on the repository front page, there are some requirements, including access to Azure Open AI, Python and some specific Python libraries.
As I continue down this road, I’ll share all the other cool things I find and create. I’m working now to convert the command-line bot to a web page to forgo the question.json for keyboard entry and then eventually a formal Azure App Service.
[Want to discuss this further? Hit me up on Twitter or LinkedIn]
[Subscribe to the RSS feed for this blog]
[Subscribe to the Weekly Microsoft Sentinel Newsletter]
[Subscribe to the Weekly Microsoft Defender Newsletter]
[Learn KQL with the Must Learn KQL series and book]