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
Because I guess the the relationship, I suppose, between the loss and the actual usefulness is super nonlinear.
I guess it's probably got to be some sort of S curve where I mean, classically with the with the self driving cars, you know, 99% accurate self driving cars.
is uh is just a heap of junk, basically, from a practical point of view.
And you maybe have to get to d ninety nine point nine nine nine.
And you don't know how many nines do you need before actually they're gonna hit the roads.
Let's push on and talk about the um the frontier safety framework, which I guess is uh Google DeepMind or uh Alphabet's equivalent of the responsible scaling policy that uh Anthropic has or the um was it preparedness framework that that OpenAI has.
These are all kind of a similar family, uh sim si similar general approach to to safety.
Um the frontier safety framework at this point is is just six pages and it's kind of a promise to write a full s uh oh well, a full suite of if-then commitments uh
next year, um, which uh you know, I ahead of releasing models that would really be jutting up against some of these some of these dangerous abilities.
When I spoke with Nick Joseph from Anthropic uh a couple of months ago now, uh, the thing that I pushed him on a bit was can is it really responsible to leave the the with the question of how dangerous these models are to the company that is developing them to decide?
Uh obviously they have an enormous conflict of interest or enormous challenge in evaluating faithfully whether the product that they're has sufficiently dangerous that maybe they shouldn't release it and start selling it.
that they I I ideally, surely you would hand it off to at least some sort of external party that that um would be able to take a more impartial perspective on it.
And to be honest, I haven't really found serious people who would say yes, what total self-regulation is uh is the future and the way that we should stick with.
Is is Google Google Do you mind taking any steps, I suppose, to
you know, uh paint a a path towards taking some of the m making some of these decisions external and taking it out of their hands to decide whether their model has passed a particular threshold of risk and particular, you know, possibly costly countermeasures are now called for.
This is slightly off off topic, but uh something that's a bit hard for front you know, the frontier safety framework or these dangerous uh capability evals to pick up is something called structural risks, which uh which actually is is is a is a name and an idea that you've been kind of a a pioneer in.
Yeah, can you just explain what what they are and like why it's a little bit hard for them to be done at the at the company level rather than the government level?
Yeah.
I guess the you know, evals might be able to pick up that a model has an ability that could feed into one of these structural risks, but because it's occurring at the societal or global level, uh, you know, it's not really possible for for GDM to be evaluating what what the entire societal effect.
I mean, it's difficult for anyone, uh, but it kinda has to be, you know, governments this is this is their remit to think about 'cause they're the ones who are best placed to to to address it.
Showing 1001–1020 of 1,787 · page 51 of 90
← Previous
Next →