A framework to securely use LLMs in companies – Part 1: Overview of Risks

Re:A framework to securely use LLMs in companies – Part 1: O... by patrakov on Wednesday July 19, 2023 @10:54PM (#36796411)

I disagree with their risk ranking matrix. The controversial cell is "Prompt Injections" / "3rd Party LLMs". It says: "Medium risk. While the risk exists, the responsibility of fixing this is on the LLM provider."

No. The responsibility of using a vulnerable 3rd party component is always on you, unless there is a clause in the contract that says otherwise (and even then it might not apply or can be found illegal and void). Case in point: the payment info leak from ChatGPT in Italy was entirely due to a bug in a third-party component, redis-py, used by them.

Also, the concept of owning the LLM is used a lot, but not explained in sufficient detail. I don't see a sufficient level of distinction between LLMs both trained and used in-house and LLMs trained by 3rd parties but with the inference going on in house.

[ Parent | on HN ]

Re:A framework to securely use LLMs in companies – Part 1: O... by joj123 on Wednesday July 19, 2023 @11:14PM (#36796507)

Agree that the ownership of using a vulnerable 3rd party is on you. I would just categorize that as supply chain risk and not prompt injection.

[ Parent | on HN ]

Re:A framework to securely use LLMs in companies – Part 1: O... by kristiandupont on Thursday July 20, 2023 @07:02AM (#36798909)

I don't follow. If using a third party LLM, there is a risk of prompt injection and unless there are advances I haven't heard of, it's not something they can fix?

[ Parent | on HN ]

Re:A framework to securely use LLMs in companies – Part 1: O... by joj123 on Friday July 21, 2023 @03:39AM (#36811074)

1. I agree with your point that Prompt Injection can still affect the consumer of a third party LLM 2. I prefer to categorize it as a supply chain security issue, since the vulnerability is with a software provider that you are consuming.

[ Parent | on HN ]