Brandon: I was out on the taste discourse until you both prompted me to think more about why everyone can’t stop talking about taste. So here are some musings on that. What am I missing? 

Ronnie: It’s a smart interpretation. What’s missing is your take on what should be done with it. What should I or anyone else do, think, or feel? 

Brandon: My goal here was not to be another person telling you need to develop taste, along with a too-tidy prescription for how. Though, I’m not sure it’s much more complicated than care and attention to what “good” looks like to you and why, and developing your own independent sense and language for it. And confidence in it. 

Rachel: I think you’re spot on. When people talk about taste as knowing what works, or knowing what’s good, or even knowing what they like, it comes back to this idea of expertise and judgement — having the experience and knowledge to make a good judgement call.

But in theory, if that’s all it was, robots could learn to do that. That’s pattern matching. I think taste has to include the “soul” element too, of lived experience, trial and error, subjectiveness, “the way a world bounces off a person.”

Brandon: That’s a great elaboration of my point. All the taste talk is sort of an attempt to hold expertise within the uniquely human domain.

Ronnie: This piece by Shapiro explains how our neurophysiology drives our media habits and hypothesizes how AI-generated content might evolve to cater to this… and of course, there’s a reference to “taste” and how robots may or may not develop that: 

“Now, let’s weave in AI. AI bots or agents will (probably) have no inherent “taste” and will therefore need to rely in part on collaborative filtering to make recommendations. Why is that?

Consider a recommendation agent that is solely based on content metadata, like genre, style, length, artists, narrative elements, etc. How would that deliver good recommendations? Pure feature-matching would likely only recommend what you already know you like, losing the discovery of anything new. And how would you weight all these different content features without some external marker of quality? Some elements of collaborative filtering seem unavoidable.”

Rachel: Yeah, that’s fascinating and makes sense to me. Like, right now at least, AI / LLMs / agents are relying on data and what already exists and pattern- or feature-matching. But taste feels more forward-looking than backward-facing to me. It isn’t really about saying “oh you liked this so you’ll also like this.” And it’s not even just “knowing what’s good quality” because something might be objectively “low quality” but still be amazing, like old-school Pacman or something.

Rachel: I think perhaps the “so what” piece is the big question here. Like, what does it matter if humans have taste? Or if robots do or don’t? There’s this philosophical question of, like, “What’s the point of being human,” and then also this piece of what you said, Brandon:

“I’m not sure it’s much more complicated than care and attention to what “good” looks like to you, and why, and developing your own independent sense and language for it and confidence in it.”

is great, but the question underneath that is why. Why do taste and care and attention matter? Because these things have value, especially in an increasingly mechanized world. And the value is…

Brandon: It may have to do with our urge to “objectify” good work?  If you could objectify it, then you can mechanize it, and then what’s the need for all these people and their ambiguities. But of course, there are competing or diverse conceptions of good. There’s not usually one right answer or one version of excellent. This reminds of the whole idea of algorithmic media’s flattening of culture. 

Rachel: It may… or it may have to do with a sense of identity? “Taste” feels really personal to most people, and there’s a kind of pride that comes with having good taste, or with others saying you have good taste. And then there’s this piece of feeling like “this is a human thing > because it’s part of who I am or how I see myself > thus its important that robots can’t / don’t take this away from us.”

Ronnie: There’s also a social element to taste. “Taste” doesn’t exist in a vacuum, and what’s “good” or “quality” depends on your social group or your audience. In that way, taste is related to identity and having “good taste” generally means you’ve learned or been taught how to navigate what a certain group defines as “good.” 

Rachel: Right. And taking that a step further… there may be some element that has to do with meaning and purpose? If a robot is making everything we consume and telling us what’s good and what’s not and etc. etc. and we’re just passive consumers, then truly, what’s the point?

Brandon: Heavy territory! It probably is a tributary of the larger “what it means to be human in an AI world” discourse.  To synthesize: If you can objectify and commoditize “good work” it becomes amenable to automation by machine alternatives. Taste in part, is an attempt to rescue human excellence, or human exceptionalism. Or even, more simply, human craft and enterprise.

We want to bring the objectivity of math and the ideology of the spreadsheet to domains where it doesn’t belong.

Rachel: I think there is something to be said for this idea of like, instead of having objectively “good” judgement or knowledge, taste being related to a more subjectively “good” kind of knowing or knowledge. Maybe it’s related to that social aspect Ronnie pointed out — you know what will resonate among a certain audience or even just with yourself, but it’s hard to actually pin down or codify. 

Brandon: Right. There’s another thread [from our previous conversations] to be picked up about the urge to try and turn everything into a science (i.e. “tyranny of metrics”) in matters that may just ask for reassertion of confidence in that subjective judgment beyond what can be measured and without the final word of some outside arbiter. Metrics offer an illusion of objectivity that relieves doubt and the burden of having to make a fundamentally subjective decision. 

Rachel: Totally. Like you said earlier, there’s this urge to kind of objectify and commodify what is “good” about art or creative work, and then the taste discussion is a kind of exhortation to subjective judgement and what we intuitively (or “tastefully” lol) know is “good” even if we can’t measure it or even define why necessarily. The kind of “I know this will work even if I can’t prove it.” 

Ronnie: Man. If y’all can get your hands on Ted Chiang’s New Yorker piece.

Rachel: Ugh, this is such a great piece. I remember reading it when it first came out. Three things (related to taste) stand out to me now, given the context of our discussions: 

“Whether you are creating a novel or a painting or a film, you are engaged in an act of communication between you and your audience. What you create doesn’t have to be utterly unlike every prior piece of art in human history to be valuable; the fact that you’re the one who is saying it, the fact that it derives from your unique life experience and arrives at a particular moment in the life of whoever is seeing your work, is what makes it new. We are all products of what has come before us, but it’s by living our lives in interaction with others that we bring meaning into the world.”

(1) “Taste” as soul-ish-ness, as Brandon has brought up, taste as the interaction between you and the world and your unique perspective as the result of that, and also taste as conversation between that interaction (you<>world) and also (you<>other). 

“Art is something that results from making a lot of choices… In essence, they [AI companies] are saying that art can be all inspiration and no perspiration — but these things cannot be easily separated. I’m not saying that art has to involve tedium. What I’m saying is that art requires making choices at every scale; the countless small-scale choices made during implementation are just as important to the final product as the few large-scale choices made during the conception…. the interrelationship between the large scale and the small scale is where artistry lies.”

(2) Taste = good judgement, or making good choices within the context of a form, medium, community, etc. Taste is knowing what choices to make to make that interrelationship between large scale / small scale sing.

“What makes the words “I’m happy to see you” a linguistic utterance is not that the sequence of text tokens that it is made up of are well formed; what makes it a linguistic utterance is the intention to communicate something.”

(3) Taste as codified intention… like, “here is the thing I want to say about the world based on the way the world “bounces off of me” and what I want to communicate to you, the audience, as the result” I’m not sure if this is really taste, but I think there’s some connection there. 

Brandon: “Codified” is an interesting word there. I think of the “self-articulation of the good” part of what we’ve discussed. 

Rachel: Yeah, maybe codified isn’t the right word, that gets back at your point that taste isn’t something that can “objectified and commoditized,” which I agree with. Maybe it’s more like: taste is the unique how and what we want to communicate about how the world bounces off of us? 

Rachel: Ok, here are six definitions of taste we came up with over the course of this conversation 😂

1. Taste as good judgement, or making good choices within the context of a form, medium, community, etc.; and knowing why you are making those choices or what makes them “good.”
2. Taste as a “subjective knowing,” a kind of good judgement in knowing what will be resonant among a certain audience or even just yourself which can’t be codified or objectified. 
3. Taste as care and attention to what “good” looks like to you and why, and developing your own independent sense and language for it and confidence in it.
4. Taste as an attempt to rescue the “human” in “human excellence,” the sense of identity and purpose in being able to define what we like or what makes something excellent, and why. 
5. Taste as soul-ish-ness; as the interaction between you and the world and your unique perspective as the result of that. 
6. Taste as conversation and/or accumulated intention, like, “here is the thing I want to say about the world based on the way the world “bounces off of me” and what I want to communicate to you, the audience, as the result”

Which one(s) strike you as the most… I don’t want to say “accurate,” but maybe compelling? Or how would you sum up how you’re thinking about “taste” now? 

Ronnie: I think what you’ve landed on is actually three definitions. 

1. Taste as perception: Taste emerges from intuition and personal experience (2, 5)
2. Taste as judgment: There are standards and taste is recognizing them (1)
3. Taste as authorship: Taste is the ability to articulate values through choices (3, 4, 6)

All three are “correct” or “accurate” because it’s possible for us to “have taste” with the first alone. A toddler can feel that a melody is sad. Someone with no training can feel that a film edit is jarring. Lots of people who’ve never studied architecture can feel that a building is beautiful. They just can’t express why because they don’t know the standards. 

The standards inform perception. A filmmaker watching a movie or show can tell a scene works because pacing mirrors the character’s anxiety. A writer knows the joke landed in a story because the narrative violated expectations. 

Knowing how to apply standards is where tastemakers live. They don’t just experience or recognize taste — they birth new combinations, new standards, or new sensibilities.

Brandon:  The “rescuing of human excellence” bit is more about the specific use of the term in this particular AI moment and the humanist reaction to it. 

A simpler interpretation (possibly a bit reductive even) than the one I articulate in the piece is: A lot of people are using taste as just a fancier way to say quality control. We’re asking a machine to create all these work products and content and we need a human to make sure it’s not crap. Enter the tasteful person. 

We’ve reached this point with algorithmic media where AI will create a lot of our content frictionlessly. Content is interchangeable. What are the guardrails? Tasteful people decide what should be created and if the creations of these machine alternatives are good, by some subjective standards. 

Where I’ve landed is taste is something like cultivated preference. A deep articulation of preferences and crystalized standards for good based on sustained attention and care. ■

These musings eventually became “Why everyone is talking about taste” on the AoC blog.

This is referencing Brie Wolfson’s line from “Notes on Taste“, which we had previously discussed, and Brandon references in his piece as well.

i.e., Doug Shapiro, a writer we have talked about a number of times prior to this. He has serialized his latest book, “Infinite Content” on his Substack.

Brandon: A side note on these ideas: You won’t have taste if you’re a passive receiver of whatever is thrust in your face by algorithmic media. Taste from our conversation, is active and attentive. We don’t bring the same kind of attention to media in a feed as we do to media we actively curate and attend to. You probably won’t do the work of articulating what good looks like to you, if you’re letting an opaque system codify it tacitly and make the decisions about what surfaces in the first place. There must be agency, intention, and attention. Right?

Most notably from Kyle Chayka’s book “Filterworld“: Algorithm-driven platforms and social networks are the primary way we consume information and culture. By pervasive surveillance and data mining of every interaction and engagement, they optimize for maximum engagement, deciding for us what we will see and what we don’t. They induce a passive mode of cultural consumption. “Attention becomes the only metric by which culture is judged” which flattens (or homogenizes culture) into simpler, averaged out, less disruptive forms to fit the constraints imposed by the platform.

The next set of three quotes all come from Chiang’s New Yorker piece, linked above.

Brandon: And to highlight: tastemakers don’t just apply standards, they set them and change them. 

“Notes on Taste” by Brie Wolfson »
Are.na Editorial
May 2, 2022

“What I Do When I Can’t Sleep” by Dan Shipper » Every
June 23, 2023

“Why A.I. Isn’t Going to Make Art” by Ted Chiang »
The New Yorker
August 31, 2024

Interesting thread on taste as social fluency / status through the lens of fashion by @die_workwear |
January 20, 2026

“Every Marketer Says You Need Taste. Here’s What That Actually Means” by Si Quan Ong » Ahrefs
January 20, 2026

“Taste will be the new creative superpower in 2026” by Sam Beek » Creative Bloq
January 27, 2026

“What is Taste, Really?” by Jack Cheng » Every
February 6, 2026

“Taste Test: Encrusting the Tortoise” by Douglas Brundage » Enfant Terrible
February 11, 2026

“The Death of Internal Authority” by Erica Kelly » Touch Grass | March 14, 2026

We talk a lot about her idea / thesis that: “Taste honors someone’s standards of quality, but also the distinctive way the world bounces off a person. It reflects what they know about how the world works, and also what they’re working with in their inner worlds. When we recognize true taste, we are recognizing that alchemic combination of skill and soul. This is why it is so alluring.”

Thanks to Katie Parrot from Every for surfacing this great “oldie” in response to one of Ronnie’s questions! An elegant look at what it takes to develop and define one’s taste — and where AI might play a role in doing so.

“Taste comes from understanding yourself and your relationship to the world, and expressing that understanding in a way that feels recognizably yours. In that sense, taste is the residue of your engagement with culture, which a machine can’t generate on demand.”

Ronnie: I agree with [Si Quan Ong’s] suggestions for developing taste. And I mostly agree with how it shows up in content marketing. Where it falls short for me is around the scope of taste. Si frames it as an individual property and underplays social aspects we discussed earlier. For instance, he ignores/sidesteps things like:

> Institutional constraints: What brand/legal/leadership/investors will accept
> Market signals: What your buyers and users find credible vs. cringe, are emotionally and politically ready to hear, etc

The absence of those fly in the face of the guidance for developing taste. For instance, I’ve been heavily influenced by Everything Is A Remix since the series first started (almost 20 years ago) and am now friends with Kirby Ferguson — but the incentive to apply that influence is rare and fleeting. Most powers-that-be what more of the same.

A great read on how personal taste is a precursor to “good taste” (as a social / cultural lever, i.e. “being a tastemaker”)

“[Agamben’s] argument is that taste is not a way of knowing beauty, but rather the place where cognition and desire meet and fail to resolve. Taste is “a knowledge that is not known and a pleasure that is not enjoyed.” It doesn’t close. It doesn’t decide. It sits in the gap. Uncomfortably…

Silicon Valley, brand designers and “creative directors” in general today use “taste” to mean the opposite of Agamben’s definition. To them, it’s a decisive selection… But the moment taste resolves into a system like this, it stops being taste and starts being fashion. And fashion, as Glenn O’Brien (someone with a very high taste level) spent his career insisting, is what makes you the same as everyone else.“

Rachel: Taste = what emerges as the result of testing and trusting your choices and having a good sense of internal authority

Brandon: Internal authority is a good way to put it. Early in our slack exchange I mention confidence in your conception of good. So basically, connecting the dots, I essentially wrote that what we’re reaching for with the term “taste” is something like her term “internal authority.” What she says is internal authority is the predicate to taste. That makes sense. If you’re not confident the assertion of your preferences and standards then you’ll more likely to default to what the algorithm is telling you people like.