A new way to build with Large Language Models

Re:A new way to build with Large Language Models by simonw on Friday February 10, 2023 @12:23AM (#34735945)

People have hooked up shells running commands already. A few days ago someone managed to extract an OpenAI API key from an environment variable because someone was running prompt output through an eval() function: https://twitter.com/ludwig_stumpp/status/1619701277419794435

Developers who don't understand prompt injection will continue to make nasty mistakes like that.

[ Parent | on HN ]

Re:A new way to build with Large Language Models by juberti on Friday February 10, 2023 @02:14AM (#34736641)

I agree caution is needed here. We have taken a few steps:

- Rate limits are enforced to provide caps on agent and function usage.

- Execution depth is capped to prevent the LLM from getting into loops.

- Function output is sanitized to prevent corruption of LLM state.

- Functions execute in a completely separate environment from the rest of the service, including the LLM, to reduce the impact from bad functions.

Note that this doesn't entirely prevent against "; DROP TABLES"-type hacks against the implementation of the function, but that problem isn't unique to us. It may however be possible for the LLM to look at function inputs and flag overtly malicious ones.

[ Parent | on HN ]