Somewhat of a tangent -- I realized I don't really understand the basic reasoning behind current efforts on "trying to make AI stop saying naughty words". Like, what's the actual problem with an LLM producing racist or otherwise offensive content that warrants so much effort? Why don't the researchers just slap an M content label on the model and be done with it? Movie characters say naughty words all the time, are racist all the time, disembody other people in ingenious and sometimes realistic ways, and nobody cares -- so what's the difference?..
Sometimes, when the pain level of not having done a task grows too high - like just before a deadline - it'll push you to do it. But this fools people into thinking that negative consequences alone will be a motivator, so they try to psyche themselves up by thinking about how bad it would be to fail. In truth, this is only making things worse, as an increased chance of failure will increase the negative motivation that's going on.
It appears that this part confuses two aspects of negative motivation: the magnitude of the consequences vs. the chance of fa...
Thank you for the great content! One question - could you please give an example of one of the "fnords" you're referencing, or explain it in slightly more detail? I think I understand what you mean but I've read the blueprint itself, and while the first few pages sound a bit grandstanding and use the word "America" way too much for my taste, I couldn't really see what you're talking about. If I didn't read your comments I'd write the tone off as "well, I guess that's how you're supposed to talk to politicians, what do I know".
Disclaimer: I'm not a native E... (read more)