High Signal Podcasts Evidence ledger
Method
Browse
How the evidence works

Public evidence record

Alexander Meinke

Published podcast speaker

Claims
7
Episodes
1
Shows
1
Named items
0

Claim ledger

What Alexander said.

7 transcript-backed records

01 / prediction

Sometimes people think the reason why we expect scheming to arise is because of some, anthropomorphizing where we think, well humans can lie and scheme and AIs are trained on, human generated data therefore we expect that AIs might scheme.

“Sometimes people think the reason why we expect scheming to arise is because of some, anthropomorphizing where we think, well humans can lie and scheme and AIs are trained on, human generated data therefore we expect that AIs might scheme.”
Publisher
Machine Learning Street Talk

02 / belief

Although, I think, like, in principle, the the math example and drug discovery, I can totally see models being very useful in those domains without goal language being applicable.

“Although, I think, like, in principle, the the math example and drug discovery, I can totally see models being very useful in those domains without goal language being applicable.”
Publisher
Machine Learning Street Talk

03 / observation

The problem is of course AIs are getting wise to these sorts of tricks and, because because they are actually being trained to be robust to prompt injections and so on so they often realize that this is not the real grader.

“The problem is of course AIs are getting wise to these sorts of tricks and, because because they are actually being trained to be robust to prompt injections and so on so they often realize that this is not the real grader.”
Publisher
Machine Learning Street Talk

04 / prediction

Otherwise, you know, the chain of thought would just get longer and longer. And when we used to be in the pre training compute dominated era, it was sort of not that important to penalize the the CoT, but the more inference costs are important and the more post training compute gets applied, the more economic pressure there is to crank the length penalty as high as you possibly can.

“Otherwise, you know, the chain of thought would just get longer and longer. And when we used to be in the pre training compute dominated era, it was sort of not that important to penalize the the CoT, but the more inference costs are important and the more post training compute gets applied, the more economic pressure there is to crank the length penalty as high as you possibly can.”
Publisher
Machine Learning Street Talk

06 / commitment

And the line kind of gets very fuzzy if eventually you start, training on deployment data for example. So because of this we're using the word reward even for graders that apply outside of training.

“And the line kind of gets very fuzzy if eventually you start, training on deployment data for example. So because of this we're using the word reward even for graders that apply outside of training.”
Publisher
Machine Learning Street Talk

07 / preference

Personally, I find it super exciting. That's exactly why we're trying to empirically study the emergence of the risks that we're worried about, because the theoretical arguments, they basically just tell you asymptotically, at some point, you should expect this to happen.

“Personally, I find it super exciting. That's exactly why we're trying to empirically study the emergence of the risks that we're worried about, because the theoretical arguments, they basically just tell you asymptotically, at some point, you should expect this to happen.”
Publisher
Machine Learning Street Talk
Search evidence