Adding plug-ins closes an air gap that has so far prevented large language models from taking actions on a person’s behalf. “We know that the models can be jailbroken, and now we’re hooking them up to the internet so that they can potentially take actions,” Hendrycks says. “That isn’t to say that by its own volition ChatGPT is going to build bombs or something, but it makes it a lot easier to do these sorts of things.”