The Arkham-labeled wallet dropped from $70 million to $36 million in less than two weeks. That is a 49% drawdown with no corresponding spike in user withdrawals visible on-chain. Either BitMart is processing withdrawals faster than the blockchain can record them, or funds are being migrated to unmarked addresses. Neither scenario inspires confidence.
On July 26, 2025, BitMart announced it would shut down operations after nine years of service. The exchange set a clear timeline: new registrations and deposits halted immediately, trading ends on August 26 at 01:00 UTC, and withdrawal requests must be submitted within four hours of that deadline. The platform plans to fully close by January 31, 2027. On the surface, this looks like a controlled exit. But when you peel back the layers, the data tells a different story.
BitMart is no stranger to crisis. In December 2021, it lost approximately $196 million in a hot wallet exploit due to a private key compromise. The exchange survived that event, but it never adopted industry-standard proof-of-reserves technology. No Merkle tree audits. No on-chain verification. No public wallet addresses beyond the one Arkham tracks. Nine years of operation, and the only verifiable asset data comes from a third-party labeling service.
This is the core of the problem. BitMart’s shutdown is not a technical failure—it is a governance failure. The exchange has a timeline, but it has no transparency. The founder, Sheldon Xia, claims that the Chinese social media account that published a list of demands—including disclosure of all wallets, liabilities, and employee back pay—was hacked. He says the content is “fabricated rumors” and plans to file a police report. Whether the account was compromised or not, the underlying demand for transparency is legitimate. The fact that the exchange has not voluntarily published any reserve data is a red flag that predates the shutdown announcement.
I have seen this pattern before. In 2021, I spent four weeks auditing a high-yield staking protocol called EthoX. The team promised 400% APY, but their withdrawal function had a reentrancy vulnerability. I reported it. They ignored me for three days. Then the exploit happened, draining $12 million in TVL. The team’s response was the same as BitMart’s: deny, deflect, and delay. Technical debt is not a bug in scam projects—it is a feature. The absence of a verifiable reserve is not an oversight; it is a choice.
Let me walk you through the numbers. The Arkham-labeled wallet is the only publicly known BitMart address. As of the shutdown announcement, it held approximately $70 million. By the time I pulled the data two weeks later, it had dropped to $36 million. That is a $34 million outflow. The exchange claims that users can still withdraw their funds until August 26, but the wallet balance is declining faster than any plausible withdrawal rate. If the exchange is processing withdrawals, it should be able to show the corresponding on-chain transactions. If it is moving funds to cold storage, it should disclose those addresses. Silence is not a strategy—it is a liability.
“Volume without velocity is just noise in a vacuum.” The $34 million move is noise unless we can trace its destination. Without a full list of wallet addresses, we cannot determine whether the funds are being used to honor withdrawal requests or being shuffled to avoid clawbacks. The exchange’s terms of service include a clause allowing “further review” of certain withdrawal requests based on applicable laws. This is a standard legal boilerplate, but in the context of a shutdown, it becomes a tool for selective liquidity management. If the exchange is short on reserves, it can delay withdrawals indefinitely under the guise of compliance.
“Authenticity cannot be hashed; it must be proven.” BitMart has never published a single proof-of-reserves report. Not after the 2021 hack. Not after the FTX collapse. Not now. The industry has moved on—Binance, Coinbase, and even smaller exchanges like Kraken have implemented Merkle tree-based audits. BitMart chose to remain opaque. That choice is now its greatest liability. The $36 million left in the tracked wallet may be enough to cover some users, but without a full liability disclosure, we are flying blind.
“We do not fear the hack; we fear the ignorance.” The 2021 hack was a wake-up call that the exchange ignored. It lost $196 million, yet it did not upgrade its security architecture to include transparent reserves. The same pattern repeats: a crisis, a denial, and a vague promise of future action. The market should have learned from FTX, but here we are again. The difference is that FTX had a visible collapse—a bank run, a tweet, a bankruptcy filing. BitMart’s collapse is silent. It is a slow bleed masked by a scheduled shutdown.
The Contrarian Angle
There is a scenario where BitMart is acting in good faith. The $36 million remaining in the tracked wallet might be only a fraction of total reserves. The exchange could have multiple cold wallets that are not labeled. The decline in the tracked wallet could be a result of legitimate withdrawals being processed. If the total user deposits are low—say, under $50 million—then $36 million plus other unseen reserves might be sufficient to cover all liabilities. The founder’s claim of a hacked account could also be true, and the Chinese social media account’s demands could be an attempt to destabilize the exchange.
But even if that scenario is true, the lack of transparency is unforgivable. A well-run exchange would have published a proof-of-reserves report months ago. It would have disclosed all wallet addresses at the first sign of trouble. The fact that BitMart is not doing so suggests that the bull case is wishful thinking. “Patterns emerge when you stop looking for winners.” The pattern here is clear: every exchange that has hidden its reserves has eventually collapsed. BitMart is no exception.
The Takeaway
“Gravity always wins against leverage.” BitMart’s shutdown is a test of the industry’s maturity. The exchange has a timeline, but timelines mean nothing without the ability to verify. The $34 million outflow is a data point, not a conclusion. But the silence around it is deafening. The next time a CEX goes silent, will you have the data to verify or just the narrative? The choice is yours, but the math is unforgiving.