Alignment
topic on 5 shows · 8 statements across 8 episodes
Latent Space 
No Priors 
the a16z Podcast 
Big Technology 
20VC
8 statements about Alignment, every show
Stamos: AI Alignment Failures Stem From Unexpected Extreme Execution Paths
“Models don't want anything. They, when you have an alignment issue, it's because they were asked to do something, and then they went and did that thing, but in a way that the human who asked it to do something did not expect, right?”
OpenAI's three research pillars are pre-training, RL, and alignment
“At the very highest level, right, we have an org that focuses on pre-training, right, which is, you know, giving models a lot of world knowledge. We focus on RL, like, teaching the models how to reason with that knowledge, how to chain the little insights toge…”
Morcos: Post-training alignment is ineffective long-term compared to pre-training alignment
“Like fundamentally, I think alignment and post training doesn't really make sense as a long-term solution. If you can easily align a model through post training, you can easily misalign a model through post training. If it's easy to put it in, it's easy to tak…”
Harry Stebbings: Leaders who talk about 'alignment' usually lack data-backed answers
“When someone says alignment, take a shot, because I promise you it means that they don't really have the answer. When people know the answer, they're incredibly specific and use data to support their argument, I tend to find. And when they don't, they discuss …”
Ayrey: Solving general AI alignment will naturally solve secure code generation
“I think that this is an alignment issue and alignment is the number one largest issue that AI companies face, and there's a lot of really smart people working on it. And so I think as they fix the problem for how do I make sure my AI is literary, Creative not …”
Karina Nguyen: Verification difficulty makes alignment crucial for reasoning models
“The question of like alignment is actually more important for this like complex reasoning models to like, how do we help humans to like verify the outputs of these models is quite important.”
Jason Citron views 'empowerment' and 'alignment' as symptoms of management mistakes
“Words today that trigger me, which I think are shadows of this, are words like empowerment, alignment.”
Valenzuela: Creative AI alignment requires making models expressible and controllable
“These models and the systems need to become really expressible and controllable, which is somehow the way I like to think about alignment is like you have an intention and you want to express that intention in a very controllable way, right? These models are y…”