Raw LLM Responses

So, we feed them this data, they analyze it, sell it back to us. Fine. But I have issues with how fast and loose this company plays with data security. Are their employees even qualified by our national data security standards to have any access to this information? H1B visa workers typically are not permitted access to national sensitive security info. That’s limited to US citizens on a need to know basis. Do they have data access audit trails if an unauthorized user attempts access? Because at last check, they couldn’t figure out how their models worked well enough to give us a trail of data sources on dubious data generated by ChatGPT. What guarantees do we have they won’t use this data in their models sold to other countries? To other intelligence agencies who will use it against us? This seems like a way to fast-tracking intelligence to bad actors across the globe who do not have US interests at heart.