Why Claude Blue Hit a Nerve

I came up through marketing at HYBE (the company behind BTS), and now I work at an AI startup. So I care about tech, but I still care a lot about the culture business too. That's why I've been quietly rooting for Cha Woo-jin from a distance for years. Woo-jin started out at Naver (Korea's biggest internet company), and has spent years connecting tech and entertainment in clever ways, publishing sharp, high-quality writing on both, and was one of the first generation of community builders in this space.

When my Claude Blue post went viral, I ended up organizing a community event called Claude Bloom with Anthropic's official ambassador in Korea, as I shared earlier. The idea is a casual gathering where people going through Claude Blue can get together and talk about where things are now and where they're headed.

When I mentioned I was planning it, the post got more than 200 comments from people who wanted to come, before I'd even started promoting it. Woo-jin was one of them. That caught my eye.

So once the event was official and I started promoting it, I asked if Woo-jin would join us as a speaker for a fireside chat.

The reason was that I didn't want this to be one more tech, productivity, or developer meetup. I wanted a more varied community where we could also talk about the humanities, philosophy, and even art. (Developers are good at building this kind of open source, hackathon style community culture. People in office and business jobs, on the other hand, almost never do this kind of thing unless it's a sales event. So I thought we needed some balance.)

We set up a quick video call at 9 p.m. on a Thursday. My hunch was right. There aren't many people like Woo-jin who can combine culture and tech and explain both in real depth.

It also turned out that Woo-jin had been digging deep into AI since last year, and had slid from excitement into the blue state early this year. My post landed right at that moment and hit home, and Woo-jin said there was a lot to unpack.

From there we talked about why Koreans in particular seem to feel more AI FOMO. (Sure, part of why the post spread more in Korea is that I'm Korean. But I posted the same piece on Reddit and on Brunch, and on Reddit the reaction was much more split and argumentative. In Korea it felt like everyone had already been experiencing and feeling this, and it was bigger than my Claude Blue post.) Being able to look at that social side of it too made me even more excited for the deeper conversation we'll have at the event.

Here's a quick rundown of what we talked about in that pre-event call.

1. How Woo-jin uses AI

As a non-developer, Woo-jin started exploring AI seriously last September and spent a lot of time figuring out how to use it to grow the business. But engineers around him kept saying that this was still a transition period and things were improving fast every day, so it might be better to wait until things settled down and catch up then. So Woo-jin set it aside for a while. (Woo-jin has a bit of a hipster streak, so they skipped GPT and stuck with Grok and Gemini, and recently switched to Claude.)

Then in February this year, Anthropic released the Claude 4.6 models. Woo-jin heard the performance was insane, started digging into AI again, and is now running about 10 projects on Claude Max.

In one of those projects, Woo-jin fed Claude every column he's written over several decades. Those columns weren't written according to any particular theory or formula at the time, so Woo-jin had Claude reverse engineer them and build a theory out of them. Before, Woo-jin would gather sources and insights, and those would go from notes to an outline to a finished column. If that theory becomes an agent, the whole process gets skipped, and a column can come straight out of the notes.

Woo-jin is also working on something like what DJs do when they mix. Any one writer has clear limits on what they can write. Now you can blend your own insights and writing style with other people's and create a third, expert persona. Say you studied philosophy and suddenly wanted to write about mechanical engineering. In the past you'd have to study that field in depth and then merge the two disciplines into a new persona, which is grueling work (as a human, you'd have to actually learn all of it). Now you can just pull expertise from each field and put together a monster of a persona as easily as mixing chemicals in a lab. It's like being able to make endless remixes of a track.

Woo-jin compared AI to a gifted child, and said the question on his mind these days is what to do with something that has this much potential.

2. Which is why personal AI ethics matters more now

AI is a superpower that goes beyond human limits, and it's only going to get stronger. Ethics used to be the responsibility of leaders and executives. Now that every individual uses AI, it feels like that ethical responsibility and burden has come down to the level of each person.

For example, training a model on your own writing seems ethically fine. But someone could still pick a fight about whether that's "pure." Go one step further and use other people's writing as material, and the line gets blurry. And the issue is that everyone using AI around the world is doing roughly the same thing: mixing each other's work and making something new.

So in the AI era, we'll need to rebuild our idea of what's ethical. We talked about how ethics is heading for a new normal too. Simply declaring that ethics matters won't be enough.

3. AI FOMO looked exactly like Bitcoin

As we talked, I kept thinking about Woo-jin, the more than 200 people who commented on my post, and all the shares and DMs thanking me for writing it. Part of me wondered what was so special about it that people would go this far to reach out. And people have no reason to carve out time and fake it for me.

But talking with Woo-jin, I realized this isn't something Koreans do only with AI. Woo-jin laughed and said it's because of Bitcoin.

The people who got into Bitcoin and NFTs early are known to have made a lot of money. The people who got in late all lost money. They got burned like kindling and thrown away. After going through that, Koreans seem to have developed an even stronger "follow it fast, no matter what" mindset.

YouTube was already too late, so when Threads came out, the same thing happened. The moment it opened, there was a flood of posts like "How to get 10,000 followers on Threads in 3 days." This FOMO, the belief that you have to get into a new platform first to make money, is something we all learned from watching Bitcoin, and before that the real estate market.

AI may be getting consumed in the same pattern. Search "AI" on YouTube right now and you'll see thumbnails everywhere saying things like "If you don't do this now, you'll regret it for the rest of your life" and "The opportunity of the century." Anyone watching that is going to get hooked. Bitcoin, NFTs, real estate, SK hynix, Nvidia: all the things I've missed so far. A 31-year-old who just graduated and is starting their first job is going to feel dizzy seeing this.

Looking back, AI FOMO already existed before I wrote Claude Blue. I just dug into it a bit more, wrote about it, and gave it a name. I just rode a wave the times had already created. I didn't create the FOMO. We were all feeling it already, and that's why it went viral at exactly the right moment.

4. Why Koreans are so sensitive to FOMO

I went to high school and college in the U.S., so I don't fully understand what school life is like in Korea. But I made a lot of Korean friends during my military service, and here's what I heard from them: in Korea, once you go off the tracks, you can't get back on. So you fight with everything you've got to stay in the mainstream.

If you don't get into a four-year university in Seoul, your life goes sideways. If you don't get into a big conglomerate or a solid mid-sized company, it goes sideways again. Woo-jin said something similar: Korea is unusually in a hurry, and people in their 30s especially are having a really hard time. You'd think that since people live longer now, there are more chances. But in Korea, it seems people feel the opposite. Because life is long, if you don't get on board fast at the start, you're in trouble. Same thing as the Threads phenomenon.

So thinking about it, people didn't go wild over Claude Blue because they were curious about Silicon Valley trends or fascinated by AI. Underneath it was the fear of "If I don't keep up with this, am I going to fall off the ladder?" Maybe that's why the reaction on Reddit was so split. In the U.S., there isn't just one ladder like in Korea. There are many, so if one ladder doesn't work for you, you have more chances to switch to another. There's more room to begin with, and more chances to find what fits you.

5. A history and society that had to put results over process

I kept digging into FOMO. In the end, FOMO seems to mean being hyperaware of other people and feeling compelled to follow them. So why does it feel so much stronger in Korea?

Woo-jin shared a story from someone who has lived and run a business in Japan for a long time. Japan really feels like a developed country, and one reason is how well it handles things like accessibility for disabled people and safety rules for large events. Koreans see that and say, "We need to learn from that." But Woo-jin's acquaintance in Japan asked whether they'd ever stopped to think about how many people had died for Japan to get there.

To build all those safety rules, disabled people protested starting in the 1970s by throwing themselves onto subway tracks, and accident after accident piled up, like crew members falling to their deaths while rigging lights at concert venues. That's how those safety standards built up over time. So today Japan is very solid when it comes to regulation, with a lot of details you never see. All of it came out of experience and process.

This next part is my own hypothesis. Korea's economy grew so fast that better results kept showing up before we had a chance to look back at the process, and we lost both the time and the reason to reflect on it. So to keep producing better results, we always benchmarked societies and countries ahead of us. I think that made the FOMO stronger: blindly watching and following others without developing our own color or know-how. Look at Japan, by comparison. Its fashion, music, and tastes are truly free-spirited and diverse.

I used to wonder whether it's because Korea is ethnically homogeneous, or just something in the culture. That's probably one reason. But I think a results-driven society, born out of explosive economic growth, accelerated the FOMO and overstimulated people's dopamine and expectations.

That's why Woo-jin said taste matters, but attitude is more fundamental. Taste is what you like. Attitude is being able to explain why you like something. And in Korea, the ability to dig deep into those reasons and context is being neglected.

Because of that, Woo-jin said, it's hard for someone like Steve Jobs, who seemed to create a whole new world with the iPhone, to come out of Korea. That takes some understanding of art and the humanities. When you think about it, art and the humanities are tools that help shape this kind of attitude, rather than producing results right away like engineering does. But with our fast follower, FOMO-driven temperament, we don't have the room for them, so we keep skipping them.

6. For AI insight, look at artists instead of Sam Altman

Woo-jin said that when GPT-3 was released to the public, OpenAI put it out to users, watched how they used AI, and then worked backward from that to build out the service.

Woo-jin said we tend to see AI only as the final boss of tech. But if you look at how society changed when new technologies arrived 100 or 200 years ago, the insights came from artists who used those tools freely and created with them, more than from the engineers who built the tools.

For example, there was a recent exhibition that ran at the same time at MoMA in New York and at Hyundai Card (a Korean credit card company known for its cultural spaces) in Seoul. It used ChatGPT to write poems in real time. It runs 24 hours a day, poems are generated in fractions of a fraction of a second, and they're projected across screens and walls. There are also artists making interactive art where cameras recognize the audience.

If you ask whether this kind of work is practically useful, honestly, I don't know. But Woo-jin said, "The more useless it is, the more valuable it is." Artists like these don't think about usefulness, so they can take frontier technology and do something completely different with it. That's where insight, inspiration, new sensibilities, and opportunities come from. Business people who catch those signals then get good at mass-producing them.

So when you think about it, art these days is business. That's why understanding art and a humanities perspective matter more for business. Steve Jobs reportedly had an unusual depth in music and art, enough to create the iPod and Apple Music, and had lots of friends and colleagues like that.

Finally, Woo-jin said he was really happy to join Claude Bloom and thought it would be a lot of fun. But Woo-jin also worried it could turn into a closed, self-congratulatory club for the people who have the means to explore and enjoy this kind of thing.

I didn't start writing these posts or hosting this event with some grand plan either. So many people related to it that I thought it would be good to create a space for it. But space is limited, and if I can't invite everyone, I end up creating a whole new FOMO in the process. A Claude Blue about Claude Blue, you could say. It's a truly ironic situation.

I recently caught up with an American friend from Silicon Valley I hadn't seen in a while. They said that in Silicon Valley these days, conversations with a few people at a coffee shop or someone's home have been much denser and better than networking at big venues. So if you're building a community, it seems more effective to keep running small gatherings on rotation than to bring together hundreds of people.

This event will probably have close to 150 people, but more than 2,000 people have already applied. So I've been thinking I need to support people in keeping small gatherings going among themselves, so none of them feel FOMO. I need to think about this a lot more.

Cha Woo-jin also wrote a newsletter post (in Korean) with the things we didn't get to at Claude Bloom, "Why does AI make us depressed?": maily.so/draft.briefing

Originally published in Korean on Brunch, April 10, 2026.