Jul 9, 2026 · 1 hr 29 min · 13 segments
What if the safest AI models weren't built by adding guardrails after training, but by shaping what gets learned in the first place? Ethan Roland, senior alignment researcher at AE Studio and first…
Ethan RolandGuest
Jacob HamesHost
And I just, what I'm curious about that aspect is, how has that shaped your perspective and thinking when it comes to AI development and AI safety? Or has it?
might be a bit you know i think i've always been really interested in chinese culture and and you know society more generally um probably one of the motivations for why i studied the language for so long but yeah i'd say that i have great respect for chinese researchers and i think that perhaps compared to the average person that hasn't engaged with Chinese culture as much that I maybe am more willing to believe in the competence and believe in the capabilities of Chinese labs.
And so I would say that, yeah, I think oftentimes the endogenous research abilities of Chinese researchers are many times underestimated.
They don't do very much alignment research in general, but I think that's more of just like a kind of what's popular versus not popular.
But in terms of just like pure research, yeah, they have their stuff together.
And there is also an argument that they actually do a good bit of alignment research.
Because, you know, they're working on actually problems similar to, like, the data filtering problem, right? They do a lot of work on unlearning, if I remember correctly.
And I think that's usually the framing in China or Chinese publications that I've read, which when you think about AI controllability techniques versus AI alignment techniques, For many things, these are very much the same thing, just like a marketing decision as to how you talk about it.
Now, that's certainly not always the case, but I think for a very large percentage of research, you could call them controllability.
And I just, what I'm curious about that aspect is, how has that shaped your perspective and thinking when it comes to AI development and AI safety? Or has it?
might be a bit you know i think i've always been really interested in chinese culture and and you know society more generally um probably one of the motivations for why i studied the language for so long but yeah i'd say that i have great respect for chinese researchers and i think that perhaps compared to the average person that hasn't engaged with Chinese culture as much that I maybe am more willing to believe in the competence and believe in the capabilities of Chinese labs.
And so I would say that, yeah, I think oftentimes the endogenous research abilities of Chinese researchers are many times underestimated.
They don't do very much alignment research in general, but I think that's more of just like a kind of what's popular versus not popular.
But in terms of just like pure research, yeah, they have their stuff together.
And there is also an argument that they actually do a good bit of alignment research.
Because, you know, they're working on actually problems similar to, like, the data filtering problem, right? They do a lot of work on unlearning, if I remember correctly.
And I think that's usually the framing in China or Chinese publications that I've read, which when you think about AI controllability techniques versus AI alignment techniques, For many things, these are very much the same thing, just like a marketing decision as to how you talk about it.
Now, that's certainly not always the case, but I think for a very large percentage of research, you could call them controllability.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
