Claim◆Video · 7:01 · 2m
Salesforce's core value isn't its CRM software—it's the deep trust that huge banks and enterprises place in it to hold their customer data, and that trust can't be replicated by a vibe-coded app.
Avishai · 20VC
ClaimVideo · 87:40 — 89:10
Because the J lens reveals what a model is representing beyond noisy chain-of-thought output, it increases trust in the model, and increased trust is what allows humans to hand over more capability and more tasks to AI systems.
Pash argues the J lens gives a clearer, more deterministic window into a model's higher-level processing than chain-of-thought alone, and that this added trust is what lets humans safely delegate more capability to AI. ✦ AI generated
Pash · The Cognitive Revolution · 2026-07-07 · original ↗
starts at this moment · 87:40
this also increases capability because one of the things about capability is you don't give capability if you don't trust it. So by increasing trust by because you have greater window and insight into what's happening you basically allow humanity to start handing over more capabilities and more tasks to these things.
verbatim transcript · starts at 87:40
Transcript · around this moment
87:40our design capabilities in a sense because now you can have more certainty. It's still not 100% certainty, right? It's still as you pointed out some at at some points is like 40% 60% 70%. It's it's still not 100% deterministic. It's still not 100% certain, but you can kind of get a greater certainty and more confidence that the model isn't hiding something from you. So in a sense, this also
88:10increases capability because one of the things about capability is you don't give capability if you don't trust it. So by increasing trust by because you have greater window and insight into what's happening you basically allow humanity to start handing over more capabilities and more tasks to these things. >> Yeah. Yeah, I suppose one way to think about it is just like how much space is is there in there to
88:46hide? And I do feel like we've got now several different ways to do like pretty meaningful monitoring that intuitively feels to me like we're we're taking enough kind of and of course there could still be more in the future and I'm sure there will be and there should be but it does feel Like we are now we've got to the point now where we've got like several different
89:22angles that make pretty incisive cuts through the model and kind of get at what is it representing, what is it thinking in different ways. >> Yeah. >> And the more of these that you kind of, you know, it's I sort of have this visual of like the old magic trick of the guy going into a barrel and then they put like a ton of swords, you know,
89:40through the barrel and it's like, well, one of those swords had to hit him, right? because there's like no nowhere left to be in that barrel with all those swords going through. I kind of feel like we're doing a similar thing with trying to understand what's going on in these models and they're not none of these things are perfect, but you put enough of these like interpretability
Related moments
Data◆Article · 44 words
Claude has a global-workspace-like internal structure — a small subset of activations called 'J-space' — that acts as a privileged representational substrate available for report, modulation, and flexible reasoning, distinct from simple chain-of-thought extraction.
Anthropic · Latent Space
Claim◆Video · 22:55 · 2m
Anthropomorphizing AI systems by asking whether they 'understand' is unnecessary, inappropriate, and a distraction — what matters is that a system's input-output behavior is predictable and useful, not whether it possesses understanding.
Michael I. Jordan · Machine Learning Street Talk