The AI Record
if you take something like status, it's like, if you are a engineer you may want to know exactly what your agent has done and you can find that in Muse. But if you're a normal person you maybe just want to know like is it doing something. So at the top it'll tell you it's working. … and then your next question might be what's it working on, you know? And so it will tell you that with the little emoji like researching pumpkin patches, but then if you want to go a little bit further you tap on it and then you have a list of everything that it's done. If you want to go a little bit further you could tap on that, and then eventually you get down to where you're going to see the curl and all the like all … and everything, which is really important from like an auditing perspective. But it's only there if you need it, and it doesn't get in the way for people who may not know what that means.

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is describing how Muse shows what the agent is doing, in answer to the host's question about the primitives the app elevates.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 07:41. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025

Filed under

Who is speaking

Alex Cornell, Design director on Muse, Meta. Leads design on the product he describes, so he sees its design decisions first-hand; he is also promoting the product, and the episode is a guest interview, so what he says about how it is used is his own account and nothing measured.

How this was checked

Checked against the source by a first reader, then confirmed independently by a second reader who fetched the source without seeing the first reader's notes (Independent review agent (Claude Sonnet), blind second pass from passage text and citation only, no repository access, 2026-11-08).

All segments match transcripts/4TOLm1g-gCE/whisper.txt after normalizing case and punctuation. The segments are 'if you take something like status', the 'normal person' sentence, the 'curl and all the' passage, and the auditing sentence. They sit near 07:43 in whisper.srt. No captions.md exists. Whisper has no speaker labels, but the intro names the guest as Alex Cornell, and he is the one explaining the app's design.

How to cite this passage

Alex Cornell (Meta), “Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo)”, How I AI (YouTube), 2026-10-12, at 07:41. https://www.youtube.com/watch?v=4TOLm1g-gCE Archived at https://the-ai-record.com/passages/meta-engineer-wants-exactly-what-agent-did-normal-person-is-it-doing-2026

More from the same source

exposing some of those additional capabilities just with the additional surfaces that we have like ideas in particular it's serving a very specific purpose which is basically just telling you some things that you may not know that it can do. I think like to your point if those were just generic suggestions that would be one thing, maybe even useful things, but when they're specifically based on what you've talked about I think that's when people really find the power

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is explaining why Muse has an ideas surface alongside the chat, in answer to the host's question about the app's primitives.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 13:12. Reported by Claire Vo.

Not yet checked [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

I think these days there's it's definitely a little bit more on the rails in terms of like where we can apply it. The avatar is a great example. And I think people find a lot of delight I have seen when they do make their own new avatar … it will then be translated into each one of those states, you know, typing on that's a little computer or making something. And that's been one of the most fun things. … if you turn yours into a tomato, then you'll see a tomato typing on the computer. And that really does feel very personal

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is responding to the host's description of generative UI in Muse, where the app's core elements stay fixed and parts inside them are generated.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 12:12. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

one of my favorite things I think to do with these systems is take what would be otherwise like a fairly mundane set of data and render it in a way that I personally find really fun. … what you're looking at here is Jolly Bot Sky Mall, which is that same set of suggestions that have looked at everything I've saved on Instagram, considered some ideas for things that I might be interested in, but then rendered them in the kind of the vernacular of the Sky Mall Sharper Image magazine.

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is screen-sharing a page his Muse generated from suggestions based on his Instagram saves.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 16:02. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

I think one thing that I have found for myself and any designer kind of coming into this project or really just this space is really wrapping your head around what a non-deterministic system is and how, what that means for when you're designing a screen. … when you first come into this space I think really understanding that these systems usually do like kind of end up the way you think … my Muse setup is completely different than yours is completely different to somebody else's. They're so personal and they can like do so many different things that you really kind of have to think more about the underlying foundation and the system, the architecture, the scaffolding that's in place and less about like, well, what about this one little thing here? Is it going to be just so

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is answering the host's lightning-round question about what designers new to agent products had to learn.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 32:15. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

the way [Muse] is designed is there's a canonical main chat that is kind of very intentionally the primary main line, and we sort of feel like you could use that entirely if you wanted to. We know that for a lot of people, especially anyone that's used to the way most of these apps work today, that threading is a very useful way to organize. So we have side chats for that purpose. Like if you want to put something specifically in, like just to get it out of your main chat or it's like a very scoped problem for … there's lots of reasons why you might want to do it. That's why side chats are there, but it is definitely like a single thread with the possibility of side threads, whereas most apps are all threaded. That's kind of like the traditional model was like a new thread every time you open it, you know? And I think once the models were able to handle the context and be able to manage everything in a single chat, to me that is a little bit easier way to interface with what you feel like … the same way you would have with another person, you know

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is answering the host's question whether to use the main chat or side chats in Muse.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 26:02. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

the basic framing that always we went back to and I think was probably the most helpful was just imagining that this interface was more of an exchange between you and an entity. And that's why it's embodied at the top of the screen. And so a lot of the patterns that we looked at were much more messaging patterns, you know, and not thinking about how did this work before … different AI apps or using kind of like the traditional industry I guess as a reference, it was more like how would you handle this if you were actually speaking with another person, you know, and that was immensely clarifying I think in almost every decision that we had to make.

Alex CornellDesign director on Muse, Meta · Leader, Meta

Context. Cornell is answering the host's question about how his team approached designing the Muse agent.

Podcast: Meta Muse: the AI agent people actually love (How I AI, hosted by Claire Vo), How I AI (YouTube), October 12, 2026, at 04:32. Reported by Claire Vo.

Checked against the source [ Grade C ] Said in the last year Meta: $201 billion revenue, FY2025 Passage page

Something wrong? Report a correction. JSON: /api/v1/passages/meta-engineer-wants-exactly-what-agent-did-normal-person-is-it-doing-2026.json