Sander Schulhoff: It does. Lenny Rachitsky: What is the issue? Sander Schulhoff: Yeah. So there's two issues here. The first one is those automated red teaming systems are always going to find something against any model. There's thousands of automated red teaming systems out there. Many of them are open source. And because all, I guess for the most part, all currently deployed chatbots are based on transformers or transformer adjacent technologies, they're all vulnerable to prompt injection gel breaking forms of adversarial attacks. And the other kind of silly thing is that when you build an automated red teaming system, you often test it on open AI models, anthropic momentals, Google models. And then when enterprises go to deploy AI systems, they're not building their own AIs for the most part. They're just grabbing one off the shelf. And so, these automated red teaming systems are not showing anything novel. It's plainly obvious to anyone that knows what they're talking about that these models can be tricked into saying whatever very easily.
And the other kind of silly thing is that when you build an automated red teaming system, you often test it on open AI models, anthropic momentals, Google models.