SEOUL, 30 AUG 2026 — Wrtn Technologies raised about 100bn won in a Series C that values it above 1 trillion won, which the company says makes it the first Korean AI service startup past that mark. The same round has been reported at three different dollar valuations.
What was raised, and by whom
The round is roughly 100bn won, about US$72m. New investors Coreline Ventures and Eugene Asset Management joined, alongside existing backers Goodwater Capital, Antler Global, Woori Venture Partners and the Korea Development Bank. Total funding to date is 230bn won.
The substance is in the growth figures. Wrtn expects revenue to clear 200bn won this year, up from 47.1bn won last year, a little over four times. Its AI entertainment platform OOC launched in North America in May and passed 10bn won in monthly revenue within three months.
The presence of the Korea Development Bank on the cap table is itself a policy signal. State-backed institutional participation in a consumer AI company is not only an investment.
One won figure, three dollar figures
The valuation has been reported as over US$722m, over US$760m and about US$870m, for the same round.
The divergence is not error so much as method. A won-denominated valuation converts at whatever rate an outlet used on whatever day, and 1 trillion won spans a wide dollar band across even a few weeks of currency movement. Some outlets are also converting a pre-money figure and others a post-money one, and headline writers round in whichever direction reads better.
Treat the won figure as the fact and the dollar figures as approximations of it. Comparing a Korean company's dollar valuation with an American one's, on numbers converted at different times, is a comparison with a currency assumption buried inside it.
The same caution applies to the milestone. The claim — first Korean AI service startup past 1 trillion won — holds up only because the category is defined so narrowly. It excludes AI generally, and the rest of Korea's technology sector. That is normal in company announcements and it is worth reading the qualifier rather than the headline.
What OOC actually is
The platform is described as AI entertainment content, and the description is doing some work.
OOC is an AI companion product: users interact with persistent characters. That is the category that has grown fastest in consumer AI outside of general assistants, and the reason 10bn won of monthly revenue arrived within three months of a US launch is that this category monetises through subscriptions and in-app purchases at rates general chatbots do not achieve.
The companion label matters more than the entertainment one, because it carries specific risks a regulator will notice. Companion products raise questions about minors, about emotional dependency, and about what happens to intimate conversation logs. Those are the questions a regulator will ask, and they are not asked of a video streaming service.
Wrtn is now regulated at home under a framework we have covered: Korea's AI Basic Act has been in force seven months with a maximum fine of about US$20,000. A company at a trillion-won valuation is not deterred by a twenty-thousand-dollar ceiling, which is the gap that law currently has.
Where a trillion won sits
The milestone's significance depends on the comparison, and two are worth making.
Against Korean technology generally it is modest. Samsung Electronics and SK Hynix are measured in hundreds of trillions of won, and 1 trillion is roughly the scale of a mid-sized listed company rather than a national champion. The claim is specifically about AI services, a category that barely existed in Korea three years ago.
Against global AI valuations it is small in a different way. A trillion won is under a billion dollars, in a year when Anthropic is preparing to tell IPO investors its addressable market exceeds US$30 trillion and Korean state budgets are written in hundreds of trillions of won.
What makes it notable is neither absolute figure. It is that a consumer AI product built in Korea is generating meaningful revenue in North America, which is the direction of travel most Korean software has historically failed to manage.
The revenue is growing faster than the valuation multiple suggests
Four times revenue growth in a year is the number that justifies this round, and it deserves the scrutiny that applies to any early consumer figure.
Rapid revenue growth in a companion product is easier to achieve than to sustain, because the early cohort is self-selecting and enthusiastic. What matters is retention at six and twelve months, and no such figure has been published.
The North American launch is also three months old. Monthly revenue at 10bn won says the product found an audience quickly; it says nothing yet about whether that audience stays, or about the customer acquisition cost of the next one.
Why a Korean consumer AI company is worth watching from here
Korea exports consumer digital products into Southeast Asia more successfully than almost anyone, through a distribution and cultural pathway that is already built.
A Korean AI companion platform that works in North America has an obvious second market in this region, where the same content ecosystem already has an audience and where smartphone-first consumption is the norm. That is a more plausible expansion path than a US company entering the same markets cold.
It also means the regulatory questions arrive here. Several ASEAN jurisdictions are drafting AI rules and none has settled how to treat companion products aimed at, or accessible to, minors. A fast-growing Korean entrant will reach those markets before the rules do, which has been the pattern for every consumer category before it.
