How North Korean Hackers Use Disposable Meme Coins to Launder Stolen Binance Crypto
Table of Contents
You might want to know
- How can counterfeit meme coins help move stolen cryptocurrency without creating a direct transfer between the wallets involved?
- Does routing funds through decentralized-exchange liquidity pools make the money untraceable?
Main Topic
North Korean-linked cryptocurrency launderers have reportedly begun using homemade meme coins to move a portion of the record $6.4 billion stolen from Binance. The technique takes advantage of decentralized exchanges, where users can trade digital assets through liquidity pools rather than relying on a conventional intermediary. According to blockchain analytics firm Elliptic, the approach inserts newly created tokens between wallets connected to stolen funds and wallets that collect the proceeds.
The method is designed to complicate the visible path of a transaction, not to erase it. A group first mints a new token and pairs it with a genuine asset, such as Near or a stablecoin, in a liquidity pool. Stolen funds are then used through one set of wallets to acquire the token. A separate set of wallets sells the token back into the pool and withdraws assets that may appear to have come from ordinary trading.
Because the pool sits between the wallets, the original funds do not move directly from the purchasing wallet to the receiving wallet. To an observer looking at a single transaction, the recipient may appear to be cashing out a newly created meme coin. The extra steps introduce distance between the wallets and can make the route harder to interpret quickly. However, Elliptic says this process does not make the money untraceable: the transactions remain visible onchain.
Elliptic’s research, credited to co-founder Tom Robinson, identified the tactic on Tron and Near, as well as a close variation on Solana. The clearest documented example was on Near. Between Nov. 5 and 7, the launderers created at least six homemade tokens, pairing each one with Near in its own liquidity pool.
The tokens used familiar meme-coin imagery and names. Their cartoon logos included a monkey wearing a bow, a flag-draped duck, a chimpanzee and a fox. One token featured an “Elon 2028 President” image of Elon Musk, who has no connection to the tokens. Names ranged from the crude FUK to SINU, a play on the names of dog-themed coins.
Creating each token cost about 25 Near tokens, or roughly $90. That comparatively small expense made the tokens disposable tools rather than projects intended to develop lasting value. The six tokens followed a similar pattern: each was launched, saw heavy trading within hours using stolen NEAR, and was then abandoned. The remaining liquidity was pulled on Nov. 8.
About 117,000 Near, roughly $430,000, passed through the six tokens. Much of the value subsequently flowed into stablecoins or through a cross-chain bridge. These amounts are small compared with the wider theft and the billions of dollars that remain a concern. Elliptic had previously found that more than $3 billion of the stolen crypto had yet to be laundered.
The operation’s broader pace is also notable. Elliptic reported that the launderers were creating a new wallet roughly every 25 seconds to move funds at machine speed. Frequent wallet creation and layered transactions can make the investigation more demanding, but the activity also leaves a large volume of recorded blockchain data for analysts to examine.
The meme-coin strategy is best understood as a way to buy time. It adds a temporary layer to the movement of funds, potentially making it less straightforward for investigators and compliance teams to connect the source wallet to the eventual recipient. Yet the blockchain records remain available for analysis, and the token trades, pool interactions, and subsequent conversions can still be followed.
Elliptic first observed this tactic last year while tracing the $1.5 billion Bybit theft. That theft was also attributed to the TraderTraitor group, which is identified as being behind the Binance breach. The recurrence suggests that disposable tokens can form one layer in a broader laundering process, rather than serving as a standalone solution.
The technique also carries practical limitations. A newly minted token and its associated pool generate observable activity, including token creation, trading, liquidity movements, and later conversions. Investigators can study those events together with wallet behavior and links to other assets. The use of several wallets may complicate attribution, but it does not remove the underlying records.
For compliance teams, the case is a reminder not to assume that assets arriving from a liquidity pool are legitimate simply because the immediate transaction resembles an ordinary trade. Newly created tokens, concentrated trading across a small group of wallets, and rapid movement into stablecoins or a cross-chain bridge may warrant closer review. The context and transaction history matter alongside the apparent source of a deposit.
More generally, decentralized exchanges can create complex transaction paths because pools hold assets and facilitate trades between many participants. That feature supports open, automated trading, but it can also be exploited to obscure how value moves. The same public data that enables these transactions can, however, help forensic analysts reconstruct activity when they examine the full sequence rather than a single transfer.
Key Insights Table
| Aspect | Description |
|---|---|
| Laundering method | New meme coins are paired with real assets in liquidity pools. Separate wallets buy and sell the tokens to put distance between the source of stolen funds and the wallets receiving proceeds. |
| Networks identified | Elliptic reported the tactic on Tron and Near, with a close variation on Solana. |
| Near token activity | Between Nov. 5 and 7, at least six tokens were created on Near. Each cost about 25 Near tokens, roughly $90, to mint, and the remaining liquidity was pulled on Nov. 8. |
| Amount moved | About 117,000 Near, roughly $430,000, passed through the six tokens before much of it moved into stablecoins or a cross-chain bridge. |
| Scale of outstanding funds | Elliptic had found more than $3 billion of the stolen crypto had yet to be laundered. The reported theft totaled $6.4 billion. |
| Traceability | The process can slow investigators and obscure direct wallet connections, but Elliptic says the blockchain trail remains traceable. |
| Earlier observation | Elliptic first observed the tactic last year while tracing the $1.5 billion Bybit theft, attributed to the TraderTraitor group. |
| Compliance implication | Funds from a liquidity pool should not automatically be treated as clean, especially when newly created tokens and a small number of trading wallets are involved. |
Afterwards...
The case points to the need for stronger tools that can examine transaction paths across decentralized exchanges, tokens, and blockchain networks. Investigators and compliance teams may benefit from systems that connect token creation, pool activity, wallet behavior, and cross-chain transfers into a single timeline. Such analysis can help distinguish ordinary trading from coordinated movement without treating every new token or liquidity-pool transaction as suspicious by default.
Further work is also needed on how exchanges, analytics firms, and public authorities share timely information while respecting legal responsibilities and the open nature of blockchain networks. Improved methods for identifying clusters of related wallets could help analysts respond more quickly when funds move at high speed. At the same time, clear standards and careful review are important to reduce false positives and avoid confusing unusual but legitimate market activity with criminal conduct.
Ultimately, this episode illustrates both the flexibility and the limits of blockchain-based laundering. Disposable meme coins can add complexity and delay, but they do not erase the public record. The most valuable area to explore further is how transparent transaction data can be turned into faster, more reliable detection across multiple networks.
Last edited at:2026/11/10
