Published: 4th September 2023
Last updated: 4th September 2023
0
I am not the first person to think of using ChatGPT to analyse HTTP requests for signs of delicious maliciousness, but that won't stop me having a crack and implementing the idea into my Server Dashboard. My first thought was to use it as a WAF, of sorts. I wouldn't rely on it to block traffic but it would be interesting to see what ChatGPT would make of inbound requests. Given that I wouldn't block traffic, I figured it would make a good integration into the Server Dashboard section of the site, where it can analyse all the requests to date.
I additionally thought it would be interesting to ask it why it thinks a particular request is malicious or otherwise.
Head over to the Server Dashboard and click on the robot head to see what ChatGPT thinks of my traffic.
Seen something in your logs?