Our First Global Meetup, in Singapore

More than 1,500 people applied and the seats were gone in a day. The night turned on a single reframe: stop asking which business is safe from AI, start asking which business grows because of it.

More than 1,500 people applied and the room sold out in 24 hours. The Claude Blue to Bloom story turned out to travel.

Asking "which business is safe from AI" produced no answer and a lot of dread. Asking "which business gets bigger because of AI" reopened the whole model.

When the cost of building collapses, the differentiator is knowing what users actually want and where it actually hurts. Models do not know that.

A cold DM on LinkedIn

Bloom held its first event outside Korea. The Seoul edition in April drew two thousand applications. Singapore drew more than 1,500, and every seat was gone within a day of the announcement.

The venue came from Singtel, Southeast Asia's largest telecom operator and a company with roughly the local name recognition AWS has elsewhere. There was no introduction and no shared network behind it. It started with cold outreach on LinkedIn.

What stood out was Singtel's counterproposal. They did not want a technology showcase. They wanted a conversation. The company is in the middle of a workforce transformation, redesigning how people work, how leaders lead, and how the business itself is shaped around AI. Plenty of large companies invest in AI infrastructure. Very few invest in AI culture.

The partnership with the Singapore Claude community, led by ambassador Max with help from Yier, a Googler, also began on LinkedIn.

That combination is the point. A local community where only Koreans gather has a ceiling. A room holding people from Korea, Japan, Singapore, Australia, the US, and the UK does not.

The question that reopened the model

The central moment of the night came from a founder describing his own slump.

He had been asking a natural question: which businesses are safe from AI? He could not find an answer, and looking for one made things worse. Every candidate had a version of the same problem.

So he inverted it. Which businesses get bigger because of AI?

That reframe changed what was actionable. It opened concrete redesigns of the business model itself, including moving from subscription pricing toward revenue sharing, where the model's leverage increases the upside instead of eroding the margin.

What models still do not know

The second thread was about what survives when production cost approaches zero.

If anyone can build the thing, building it is not the differentiator. Knowing what users actually want, where the workflow actually hurts, and what the domain actually punishes stays hard. That knowledge comes from proximity to real customers, and it does not arrive through a model.

The same logic applied to tooling. Chasing every new release is a fast route to burnout. The healthier pattern is to keep only a few tools per layer and evaluate anything new against a single question: where does this fit in the kit I already have?

Presence as a muscle

One speaker, who built a project called The Cave using Claude Code and Gemini, framed rumination and worry as the practice of living somewhere other than here. He talked about presence as something you train rather than something you have.

The community demos matched that spirit. A quiet invite-only social platform designed so that participation does not depend on manufacturing outrage or FOMO. A perfume bottle generator that lets anyone play at being a perfumer. Pendulum Type, which writes letterforms out of the chaos of a double pendulum.

None of them position the user as a spectator receiving output. All of them hand back the process.

What a company can take from this

Four things translate directly into practice.

Change the question to find the opening. Swap "which business is safe from AI" for "which business grows because of AI." The second version produces work you can actually do, including pricing model redesign.

Hold the part models cannot reach. As build cost falls, empathy, domain knowledge, and proximity to the customer become the differentiators rather than the soft skills.

Simplify the toolkit on purpose. Keep a few tools per layer. Judge new ones by fit, not novelty. Sustainability beats coverage.

Treat offline connection as an asset. Burnout rises alongside adoption. Rooms where people meet face to face buffer the isolation and build real support and real networks.

What "Claude Blue to Bloom" means

The first stretch of using AI is exciting. The more you use it, the more the FOMO, the job anxiety, and the identity questions accumulate, and the excitement turns blue. Nobody wants to stay there. Bloom is the name for what comes after that feeling rather than instead of it.

On the closing stage the team announced Bloom Seoul, a thousand-person global conference in Korea in late October. Filling a room that size requires people from outside Korea. Which is the honest reason this first meetup outside Seoul should not be the last one.