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
Yeah.
Um so I guess to come back to it.
What what are the ways that it's that it's more difficult for for AIs to cooperate?
almost magic.
So we also have enormous historical experience to kind of judge how do people behave in different situations.
I guess the the extreme instance of this is that they can be potentially be backdoored.
Uh so they could have, you know, you could have sort of m I I think on the c under the current technology we almost can can implant magic words into AIs that can cause them to behave completely different than the than the than they would previously.
And this is extremely hard to detect, uh at least with current levels of of interpretability.
Um so that could be something that could really trouble uh that you could have an AI completely flip behavior um in response to slightly different conditions.
This is slightly out of place.
But what I mentioned, this crazy idea that you could use the possibility of having backdoored a model to make it extremely undesirable to steal a model from someone else and apply it.
You can imagine, you know, the US might say, well, maybe we've backdoored the models that we're using in our national security infrastructure.
So that if they detect that they're uh being operated by a different country, then they're gonna completely flip out and behave incredibly differently.
And there'll be almost no way to detect that.
So that could be, I mean.
I think it's a bad situation in general, but this could be one way that you could make it more difficult to to hack an and and take advantage or just like steal at the last minute the model of a different group.
a virtuous cycle.
sounds a little bit like dancing on a knife, uh knife edge.
Um
Okay, so that's some ways that it might be more difficult for AIs to cooperate.
Showing 801–820 of 1,787 · page 41 of 90
← Previous
Next →