Seeing sporadic cases where model responses don’t load fully for some networks—anyone else experiencing similar issues?

Hi all,
I’m seeing sporadic responses from a customer network when using chatkit and agentbuilder. The response sometimes load partly and then there is an error message that it was unable to generate the response. – I have not been able to recreate the issue and everythings works completely fine for us on multiple networks (office and mobile hotspot) and tested multiple browsers (macos/windows) – does anyone else experience something similar? like enterprise anti-virus or network firewalls limiting access to openai? it doesn’t seem to be a rate limit as we have low user count and during our own testing we don’t get issues…