Rob Wiblin
speaker
1,787 appearances
3 recordings
1 series
first heard Sep 2024
last heard May 2025
Rob Wiblin’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
I mean
Maybe another reason why this research agenda doesn't feel intuitively so absolutely essential is that the AIs that I that I interact with, you know, L you know, LLMs, they just seem like very cooperative and very nice by by by nature.
And it's kind of easy to imagine scaling them up, doing the same sort of RLHF that we're doing to produce that sort of personality now, and say, well, wouldn't they continue to act to have nice personalities and be really quite cooperative by by nature?
What do you think of that?
Yeah, I mean, a nice thing about um AR models is because you because you can test them and then use exactly the same model to consider the the new inputs as all of the previous ones, you can trust them maybe as a mediator that but look by looking at the track record to say, well, they they produced fair outcomes in all these previous cases as a judge.
So um I would like trust it to produce um a fair, a fair outcome in the in the in the in in this in this new dispute that that that I'm engaged in.
So I guess yeah a lot of a lot of potential there.
What is the actual agenda for trying to make AIs more cooperative?
Are there are there people working on it?
Like what what what actual technologies do we need to develop?
Okay.
So you have lots of different tests for how effectively these AIs have cooperated.
And I suppose you can you can have uh like almost conflicting scenarios where you'd almost require different behaviors in order to produce uh a nice cooperative uh outcome in these different cases.
I guess it's like
running lots of kind of different game game theoretics scenarios that might demand that you be a hard ass in other cases and that you be friendly in other cases.
I suppose there's this there's this history of all of these game theory tournaments where trying to try to figure out what are the most simple cooperative agents that you can have.
And then I guess you can just have a research program to try to figure out to try to develop the the most uh successful uh cooperative agent that that can that can successfully produce um a lot of utility across a lot of different um yeah possible situations that might be thrown
into.
Yeah.
Um I've noticed this funny phenomenon.
Showing 821–840 of 1,787 · page 42 of 90
← Previous
Next →