[brabel]

Couldn't this open people up for remote code execution somehow? Say, someone sends you a message that they know will make you likely to ask an AI a certain question in a certain way... Maybe far-fetched, but I've seen even more far-fetched attacks in real life :D

[kolinko]

the code is sandboxed on openai servers. it doesn’t run on your machine if you use chatgpt interface

[joquarky]

I would assume it can only generate pure functions and/or run in a sandbox.