South Korean police refer 18 Polymarket users to prosecutors in gambling probe
Gangwon investigators cited 17.6B won ($12.7M) in wagers and said on-chain data was enough to identify users.
South Korean police have referred 18 Polymarket users to prosecutors in an illegal gambling investigation after tracing activity tied to about 17.6 billion won ($12.7 million) in wagers. The case marks an escalation from August’s access-blocking move into a prosecution-stage posture that traders now have to price as personal jurisdiction risk.
Key Takeaways
- Gangwon Provincial Police Agency referred 18 Polymarket users to prosecutors after investigating 26 people in an illegal gambling case.
- Investigators cited about 17.6 billion won (about $12.7 million) in collective wagers, including a maximum single-user wager of about 5.7 billion won (about $4.1 million).
- Police said publicly available blockchain transactions were sufficient to identify users tied to the activity.
- Authorities are treating Polymarket activity as illegal gambling under South Korea’s Criminal Act, while users have argued it functions more like a crypto-based derivatives investment market.
Referrals to Prosecutors Mark a New Phase in South Korea’s Polymarket Case
South Korea’s Polymarket response has moved past the “block access and warn” stage. The Gangwon Provincial Police Agency has referred 18 Polymarket users to prosecutors as part of an illegal gambling investigation involving 26 people, based on data submitted by the National Police Agency to the office of Democratic Party lawmaker Yoon Kun-young.
The procedural distinction matters. A referral does not mean indictments have been filed, and the packet does not specify what charges prosecutors may pursue or what penalties could attach, but it does move the case into the part of the system where outcomes start to harden, and where defense arguments get tested against the Criminal Act rather than debated in the abstract.
The investigation began in June 2026, when Gangwon police launched what was described as South Korea’s first illegal gambling probe into local Polymarket users at the request of the National Police Agency. On Aug. 18, authorities moved to block Polymarket after determining it provided an illegal gambling environment to users in the country. The new development is that end-users are now being sent to prosecutors, not just pushed off the front door.
The Numbers Behind the Probe: $12.7M in Wagers and a $4.1M Single-User Exposure
The disclosed figures put real size behind what could otherwise be dismissed as a symbolic action. Police cited about 17.6 billion won in collective wagers across the 26 investigated people, roughly $12.7 million, and said the largest amount wagered by a single user was about 5.7 billion won, roughly $4.1 million.
Two things stand out for traders. First, the concentration implied by a $4.1 million maximum suggests investigators are not limiting attention to small retail behavior, and the case selection likely reflects an enforcement preference for meaningful exposure that can justify investigative work and prosecutorial time. Second, the aggregate total is large enough to signal that authorities view this as more than a technical violation, even if the legal theory ultimately turns on definitions rather than size.
The legal framing is also tightening. Authorities’ position was summarized in the report as follows: “Authorities reportedly said Polymarket transactions constitute illegal gambling under South Korea’s Criminal Act because users stake assets on outcomes that cannot be predicted with certainty.” Users pushed back with a market-structure argument, with the report noting: “The users argued that Polymarket should instead be treated as a crypto-based derivatives investment market, according to the report.”
That dispute is not just semantics. If prosecutors and courts accept the gambling characterization, the “I was trading a market” defense becomes a harder sell, and the focus can shift from platform access to individual participation.
On-Chain Traceability: Police Say Public Blockchain Data Was Enough to Identify Users
Police said they identified users by analyzing publicly available blockchain transactions. That is the most operationally important detail for anyone using noncustodial products under a restrictive jurisdiction, because it undercuts the assumption that the absence of a conventional real-name user list is a practical shield.
Polymarket, as described in the packet, lets users buy and sell contracts tied to real-world outcomes and operates on a noncustodial, peer-to-peer structure with automated settlement. The report also described it as not maintaining a conventional list of users by their real names. In other words, the platform design is the kind of structure many traders associate with reduced counterparty and account risk, but this case is a reminder that noncustodial does not mean non-attributable when investigators can work from on-chain flows.
The packet does not specify which blockchain(s) were analyzed, which addresses were involved, or how investigators linked addresses to individuals beyond the claim that the transactions were publicly available. That missing methodology matters, because the practical risk to other users depends on whether identification relied on exchange off-ramps, repeated address reuse, counterparties, or other linkage points.
South Korea’s earlier blocking decision also shows where authorities think the “platform vs protocol” line sits. The country’s media and communications review commission cited a winner-takes-all structure and pointed to Polymarket’s role in operating markets, setting trading rules, providing crypto deposits, withdrawals, and settlement, and collecting transaction fees. Polymarket argued it did not provide Korean-language services or support payments in Korean won, and that noncustodial smart-contract transactions meant it did not directly manage user funds. The commission rejected that argument, with the report quoting: “The commission rejected the argument, saying technical characteristics did not exempt a service from South Korean law.”
Tae-Lim Kim, managing attorney at AXIS Law, told Asia Economy that the transactions could meet legal requirements for gambling, and said describing them as prediction derivatives would be difficult to use as a direct defense in criminal proceedings. He added that the ability to trade contracts and exit positions before settlement could still be relevant to a court’s assessment.
Signals Traders Should Track After the Aug. 18 Block and the Prosecutor Referrals
The next concrete signal is whether prosecutors file indictments or announce specific charges and penalties for any of the 18 referred users. The packet does not provide outcomes, and a referral can sit without immediate public action, but charging decisions are the point where “enforcement posture” becomes “enforcement record.”
A second signal is whether enforcement expands beyond users to intermediaries or related services following the Aug. 18 move to block Polymarket. The commission’s reasoning emphasized market operation, rule-setting, settlement, and fee collection, which can map onto a broader set of actors than just end-users.
Third, traders should pay attention to any further guidance from the media and communications review commission on which platform features trigger gambling classification, particularly the winner-takes-all structure and the operational roles it listed. That kind of feature-based test tends to travel, because it gives agencies a reusable template for the next product.
Finally, the most important unresolved detail is the linkage method. Police said public blockchain transactions were enough to identify users, but the packet does not explain how addresses were tied to real-world identities. Any new disclosure on that point will change how on-chain participants model personal risk, especially for “perps-like” products that rely on frequent position changes and settlement flows.
My Read: This Case Prices In Jurisdiction Risk for ‘Perps-Like’ On-Chain Products
The referral of 18 users is being read as just another access fight, and I don’t think that survives contact with the procedural step. Blocking a site is a policy statement. Sending users to prosecutors is a deterrence mechanism, and the disclosed wager sizes, including a 5.7 billion won single-user figure, suggest authorities are comfortable anchoring the case on meaningful exposure rather than only making an example of small accounts.
The threshold that matters is whether prosecutors turn these referrals into indictments under the Criminal Act, because that would convert South Korea’s theory, that staking on uncertain outcomes is gambling even when wrapped in a noncustodial market, into precedent that traders will have to price across similar “perps-like” on-chain products. If indictments land and the on-chain attribution method holds up in court, jurisdiction risk stops being a narrative and becomes a compliance constraint with named defendants.
