By 2030, We'll Be Debating Whether AI Is a Person

Same 2030, very different feelings

A developer friend I met one afternoon told me what they always say when they see friends whose work has nothing to do with AI.

By 2030 or so, we won't be talking about whether AI replaces us. We'll be talking about whether to treat AI as a person.

My friend thinks most of today's AI conversation is pretty shallow. Plenty of experts are too busy with their own lives to look very far ahead. And when they recently asked friends outside the AI world about it, their reaction was, "Wow, they really have no idea." They all assumed the point where AI replaces them was much further off than it is. They bring up 2030 on purpose, because they think ordinary people deserve some time to prepare. People who are better with AI than they are will prepare on their own, my friend said, but even they don't have much time to think seriously about the future after AI.

They mentioned a passage they'd read in Seneca, roughly that once you pull away from the crowd, your worries fall away too. The people who feel anxious about AI, they said, are the ones who keep hearing AI news. If you hear it every day, you get anxious. If you don't, all of it sounds like stories from the moon. Two people can look at the same year, 2030, and stand that far apart.

The people who feel it first leave first

An older friend who runs a startup told me that two friends of the most senior developer on his team quit development last week. They said they felt they had done all they could, and now they're looking for something else to do. He thinks the people who feel this shift in their bones first are the first ones to jump.

On his own team, it was the developers who first said, "Wouldn't we be better off teaching AI?" So the team decided to teach people in the neighborhood one on one, whenever someone asked, "How do I use Codex?" They could charge for it or not. Someone even suggested opening an after-school academy that teaches AI to elementary school kids. (In Korea these private academies, called hagwons, are a huge industry.) His view was that education is where the money is, so everyone will end up heading that way sooner or later.

That day at lunch, he told his developers they couldn't just code anymore. They had to be able to do anything.

Right now real content is the moat, so let's think of ourselves as a content company.

He said everyone now needs content of their own, and has to do something, anything, whether that means becoming an influencer or something else. People who just want to keep doing what they've always done will get pushed out of the market. Startups take VC money and expect a J-curve, so they often spend the early days worrying only about whether the idea will scale, and end up never doing anything at all.

You think you'll know ahead of time whether it scales? You scale it by doing it.

I also think that when you only think something through in your head, what you're really missing is the experience of trying it. Knowing something in your head means nothing if you can't act on it. I suspect the people who felt the change first are moving first because they know you only learn this by running into it yourself.

How we work and how we make money are both changing

A colleague who runs Bloom with me (Bloom is the AI community I run in Seoul) strongly agreed with something they heard at a Mixer, one of our smaller meetups: "AI will raise the productivity of non-developers much more than developers." When I prepare Bloom events, the admin work far outweighs the actual work, and I feel like AI could handle all of that. I joked that the real hands-on work is showing up an hour early and setting out chairs.

Someone I met at a lunch Mixer works in a heavily regulated industry. Because of personal data rules, it's hard to even judge whether they're allowed to automate something. In places like that, the real barrier is security. But if you think about it, we don't hand personal data to people because people are especially good at handling it. We have them sign a confidentiality agreement and then we trust them with it. That's why we put security checks on people too, like access controls on who can enter. If AI has proper safeguards, such as running on an internal network, there's no reason it can't do the job. And as models get lighter, quite a few companies are already running them inside their own networks.

At a Friday Mixer, someone pointed out that far more people are drawn to "how to make money with AI" than to "how to work faster with AI." This came from a person who had experimented by posting short-form videos. Videos about using AI to be more productive got almost no response. Videos about how to rack up views easily drew enormous crowds.

Some people are already using personal AI agents as a family. With a service like Instinct, for example, my agent reaches out to your agent first, and your agent sums things up and passes the message on to you. It works the way executives' assistants sort out a meeting time before the bosses ever talk. People use it to deliver awkward things they'd rather not say directly, in a softer way. The reaction was that it's convenient, but also a little sad.

If every day becomes a weekend

A friend in the US told me, "It's going to be true that a lot of people won't be needed." But my friend thought that could be a good thing in the end. If a small number of people can run the systems we already have more efficiently, that's a very good thing, they said. If the system can supply everything we need quickly and cheaply, then everyone can kick back and still get what they want.

I said people are social animals, and we'd resent seeing someone else loafing around. We keep comparing ourselves to each other, and we keep building hierarchies. People don't like everyone being the same. Even if we all wear similar clothes, it bothers us when someone else's are nicer. When I asked whether classes would form even in that world, my friend answered:

Not everyone has to play the same way. Everyone just gets the same opportunities, and what you pick is up to you. Every day is a weekend.

I brought up something I'd heard somewhere: that the companies at the AI frontier will all end up becoming bio companies. My friend said that probably means once we've solved all the world's problems, we go back to basic science. I told them that at the lunch Mixer people kept saying "an age of abundance is coming," but I'm not sure yet that it will actually be a good world. Between my friend, who pictures a world where every day is a weekend, and people who say "I don't even know about next month," I honestly don't know yet where I stand.

A place to connect the two

What my developer friend said about Bloom touched on the same gap. The people at the AI frontier are too busy with research to talk about the bigger picture. The people with the philosophical depth to talk about the bigger picture don't know the technology, so they can't. Someone has to bridge that and talk it through, and my friend said Bloom seems to shine because Korea has almost no place like that.

I think the best thing would be for Bloom to take over the whole AI conversation in Korea.

At the same time, my friend said the US is flooded with AI discourse. AI has become a topic everyone cares about, so it would be perfectly normal for any outlet to cover it. Seen that way, they added (and this one stung), they couldn't think of anything in the topics or format Bloom covers that would be hard for others to copy. For a community to last, there has to be at least one person obsessed with the topic, someone who will keep digging into it whether people show up or not. And a community, in the end, has to become a media outlet.

I also find it hard in Seoul to find people who can talk about life after AGI. For an event to be interesting, it needs new ideas you can't already get on YouTube, and few people can offer that. That's why my friend is preparing videos too. They think informational YouTube doesn't have much life left. They read those information videos as transcripts anyway. So instead of videos that just carry information, they want to make videos that show their process of figuring out how to survive after AGI. I added that founders abroad build a fan base by posting bits of their building process before the product even exists.

My friend in the US recommended "Range" to me, a book arguing that generalists who know a little about many fields end up succeeding more than specialists who go deep in one. They said it seemed like the right book for me, since I stay close to people from all kinds of fields. When I'm stuck inside my own industry I often can't see new ideas. But when I look at another industry and compare it to mine, my own industry sometimes comes into focus, like seeing it in a mirror.

Some people are already living in 2030. Others still hear about it like stories from the moon. I believe connecting them starts with getting both of them to sit at the same table and listen to each other. That's why I keep meeting people. As I told my older friend that evening, the coffee chats themselves are my content, and they're what grows the community.

Originally published in Korean on Brunch, October 17, 2026.