Beyond Privacy Trade-offs with Structured Transparency — mentioned by williamtrask

Sun, 05 May 2024 15:50:40 GMT

You seem to be focusing a lot on remembering or forgetting consequences. Yes, ensuring models know enough about the world to only cause the consequences they desire is a good way for models to not create random harm. This is probably a good thing. However, there are many other reasons why you might want a neural network to provably forget something. The main reason has to do with structuring an AGI's power. Even though the simple-story of AGI is something like "make it super powerful, general, and value aligned and humanity will prosper". However, the reality is more nuanced. Sometimes you…

Paper and discussion history. A mention is not an endorsement.

Beyond Privacy Trade-offs with Structured Transparency — HN submission

Thu, 11 Apr 2024 16:14:34 GMT

Structured Transparency: a framework for addressing use/mis-use trade-offs

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Beyond Privacy Trade-offs with Structured Transparency · HackerPapers discussions

Beyond Privacy Trade-offs with Structured Transparency — mentioned by williamtrask

Beyond Privacy Trade-offs with Structured Transparency — HN submission