Yeah, there are many possibilities, and I wish OpenAI were more open[1] about what went into training Bing Chat
Ok but here’s a joke you could make at this point:
Alignment researchers: Please stop posting capabilities research on Arxiv.
Also alignment researchers: My six layer prompt injection just caused the AI to act mildly aggressively. Where are my kernel dumps?!
Agreed, this seems rather unfair.
Ok but here’s a joke you could make at this point:
Agreed, this seems rather unfair.