Who Prices PUBG: Four KRAFTON Controversies and One Balance Sheet

One promise was 100 percent. The proven reality was 9 percent. In 2026 KRAFTON placed two paid products in the PUBG: Battlegrounds store — the PUBG X NewJeans Loot Pack and the PUBG X NewJeans Premium Bundle. Buyers opened them for NewJeans-inspired collectibles and character customisation. The published rule was simple: after four consecutive failures, the fifth pull would guarantee the set blueprint. Some players bought more than five Premium Bundles and never received it. The Korea Fair Trade Commission investigation found the Premium Bundle sat outside the pity system altogether, and that some information displayed in game was flatly wrong. In June 2026, Yonhap reported the commission fined KRAFTON 2.5 million won for providing misleading information about item acquisition probabilities in PUBG: Battlegrounds.

I follow the ball, but I file the balance sheet. The number that matters here is not the fine. It is 9 against 100. When a company prints a probability, it stops being marketing copy and becomes a pricing document. When that document is wrong, the company does not merely make a mistake — it leaves a blank cheque inside its own product.

To read KRAFTON you have to read the revenue architecture of a Korean game publisher. The core asset is PUBG: Battlegrounds — PC and console first, mobile second, esports ecosystem third. Four revenue layers are visible. Game sales and base in-game purchases. The cosmetics economy, where loot boxes and limited-time collaboration packs are the actual profit engine. China, where KRAFTON earns a technical service fee for Tencent-operated Peacekeeper Elite, a fact the company first admitted in a 2026 board filing. And esports: tournaments like PGC, PGS and PNC, plus sponsorship contracts and broadcast rights.

Across those four layers the company has made headlines four times in two years. From outside, they look like four unrelated scandals. To me they are four faces of one problem — the gap between the published rule and the operating rule. When a product is priced on probability, that gap is the most expensive weakness a company can carry.

The first thread comes from an entirely different direction. In 2026, Korean gaming communities alleged that some players were using NewJeans members' face skins to create and share harassing imagery, pairing faces with revealing outfits. The case became more sensitive because Haerin and Hyein were minors at the time. This is where I stop, and where the material should not be converted into a number. Two teenage faces had entered a video game's cosmetic inventory, and that inventory was never built to protect them. That cost appears in no quarterly report, and it should not.

KRAFTON and ADOR later said they would act against uses inconsistent with the collaboration's original purpose. KRAFTON subsequently restricted combining NewJeans face skins with certain outfits. Consensus formed quickly: the problem was player behaviour. My read differs. PUBG's character system holds face and body in separate inventory slots — that architecture created the exploit surface. The design review before the collaboration should have asked how the company's own system could be broken. It did not, and the collaboration paid the price.

The second thread runs back to 2026. ZDNet Korea reported that, ahead of its IPO, KRAFTON disclosed in its securities registration filing that it provides technical services for Tencent-operated Peacekeeper Elite in China and receives a service fee. That revived an older argument. PUBG Mobile shut down in China in May 2026 over licensing problems, and Tencent moved players to Peacekeeper Elite, a similar title rebuilt in gameplay, graphics and content. KRAFTON maintained the two were separate products. Some argued the differences were sufficient to meet licensing conditions; others asked whether this was a way to keep PUBG operating in China.

My read: this was never a product-definition question. It was a revenue-recognition question. The company converted a geopolitical risk into a service contract, and the driver of that conversion was not IP — it was the distribution channel. Working in Chengdu taught me that in markets which block client apps, valuation is set by service terms, not user counts. And that kind of valuation is almost always bought at a discount, because the buyer knows the asset is revocable.

Back to the 2026 loot pack case, this time on numbers alone. The commission's finding: published probability was one thing, actual probability 9 percent. KRAFTON later refunded roughly 1.1 billion won to about 380,000 buyers and issued in-game compensation valued at roughly 9.8 billion won. Because the company corrected its disclosure and paid compensation itself, the action was limited to a fine.

Two things need separating. The refund, compensation and 2.5 million won fine are observed figures, per Yonhap's June 2026 report. My analysis beyond that is a model, not an audit. Anyone pricing the in-game compensation at face value will see a different total, and that reading is legitimate.

The second thing is structural. The compensation was paid in the currency the company prints. The spread between the marginal cost of in-game currency and its face value determines the true price of misleading disclosure. 1.1 billion in cash, 9.8 billion on paper, 2.5 million in the regulator's hand. When a fine looks like a rounding error beside the remediation cost, the market receives a signal: misstate, apologise quickly, and the price is manageable. That is the moral hazard no balance sheet line captures — only competitor behaviour does.

Some will argue voluntary correction justifies a light penalty. I disagree, and my objection is procedural, not emotional. The regulator measured disclosure accuracy. What went unmeasured was time — how long players bought on false information, and what that time was worth. Time is a cost, and nobody refunded it.

Now the most recent case, and the one that says the most. At PUBG Asia Stars 2026, allegations emerged that two Vietnamese players — Himass of Anyone's Legend and TanVuu of The Expendables — were drawing on opponents' livestreams. On 23 September KRAFTON published its investigation: both had used external information, including other players' streams, to form judgements and build tactics in matches. KRAFTON classified it as stream sniping, a breach of PUBG: Battlegrounds operating policy and the conduct regulations for professional players. The penalty: permanent account closure and a ban from any KRAFTON-run or approved PUBG Esports event, including PGC, PGS and PNC.

The case began when Korean streamer Soopi of Gen.G alleged the two Vietnamese players were watching opponents' streams. Amid the fallout, organisers removed both from the rest of the tournament, adjusted scores, increased stream delay, cancelled the third match day and split the prize money evenly. KRAFTON later said it found no further violations.

But the punishment outgrew the incident. Reaction from Vietnam's PUBG community was enormous — more than 4.1 million signatures were collected in support of Himass and TanVuu. Among streamers and influencers, Độ Mixi, PewPew, Rambo, DjChip and Ngân Sát Thủ protested or removed PUBG. GAM x The Expendables demanded a clear explanation, while Anyone's Legend defended the players' right to respond. Debate split quickly: whether the players were guilty, and how proportionate and consistently applied the sanction was.

My read is blunt, and I know it is uncomfortable. When the same company misstates its own probabilities, it takes a year, a light fine and in-game compensation; when suspicion reaches its players, it takes a permanent ban within days. That speed differential is the real liability. In gaming, rules are priced on impartiality, and impartiality is measured by equality of enforcement. When the ecosystem's largest payer and its players sit at one table under two standards, sponsors reprice quietly — not in new contracts, but in new terms.

Here I owe a confession. I left Seoul for Chengdu and came back with a business model. In Moscow I learned that a World Cup has a business desk. Empty stadiums taught me the crowd is a revenue line, not just noise. In 2026, covering the Euros and Tokyo, I overruled two colleagues who wanted a softer angle, and later had to apologise for that call. That is my audited failure. So I flag my own risk here.

In the Vietnamese context my reading may be wrong. I call the process the bigger problem, but a counterpart working in Vietnamese tournament operations told me the opposite. For him the issue is not process but magnitude. His argument was practical: finishing an investigation within a match day and issuing a permanent ban means the accused never gets a full defence heard. That is a procedural objection, not a careless one.

My own defence, for the record: where competitive integrity is at stake, an operator must decide fast, because delay leaves the event's outcome undefined. I know, because I have sat in that room. But 4.1 million signatures are not a sentiment metric to me. They are a market signal. Vietnam is one of PUBG's highest-attention markets, and attention is the only stadium nobody can close. Esports taught me that attention is the real stadium.

Someone will say a signature drive is a viral moment, and viral moments are not taxable. True. But untaxable does not mean unaccounted — it sits in the risk premium. And that premium sets how fast a community buys the next pack, and how hard a third-party tournament organiser bargains.

Over the next twelve months I will watch three numbers. Probability disclosure rules, which will tighten in Korea and Europe and turn loot boxes from cosmetic products into documented, regulated ones where every rate is contestable and every change is archived. The term and renewal of the China service-fee revenue line, where valuation rests on relationship stability rather than IP independence. And the quietest one: the structure of new sponsorship and tournament contracts coming out of Vietnam.

If new terms appear there — appeal rights, investigation deadlines, disclosure obligations — that will be the company's most expensive lesson. It will show up in contract clauses, not press releases.

A cruel arithmetic exists. A 2.5 million won fine, 1.1 billion won in cash refunds, 9.8 billion won in in-game compensation: together, the remediation cost approaches the budget of a mid-size collaboration. In exchange the company bought a lesson and pushed a community further away. Which is larger depends on what you measure — quarterly profit, or three-year retention.

The answer is clear to me, because I once sat in an empty stadium and learned that even when the cameras leave, the noise does not stop. A community that stops believing your probability table cannot be brought back with fines, only with evidence. Which leaves KRAFTON one simple question before it puts the next pack on the shelf: what will it use to prove its own number?