So in a situation where you are communicating with let’s say an AI endpoint that generates same response but in a different or paraphrased sentence per api call, this will be an issue on the mainnet when making http outcalls to the server?
You can create a transformation function to remove parts of the response that would be different between calls made milliseconds of each other.