AlphaFi is shutting down its decentralized finance operations on Sui even after covering the bad debt that triggered the crisis, an unusual outcome that shifts attention from whether users can be repaid to whether the protocol remained viable after the failure.
The protocol entered maintenance mode on September 26, stopping new deposits and borrowing while leaving withdrawals open. AlphaFi has asked users to close existing positions and remove their assets and says it will formally exit the Sui ecosystem after the final user has withdrawn.
The shutdown follows severe undercollateralization in AlphaLend borrowing positions backed by ALPHA, AlphaFi’s ecosystem token. Those positions generated bad debt after the collateral supporting outstanding loans was no longer sufficient to cover what borrowers owed.
AlphaFi says the bad debt has now been fully covered, the protocol remains solvent and user funds are secure. The team is continuing to work with the Sui Foundation on the investigation and attempts to recover outstanding debt.
That makes this different from a conventional DeFi collapse. AlphaFi is not closing because it currently lacks enough assets to honor withdrawals. It is winding down after restoring solvency.
AlphaFi Has Removed Lockups to Get Users Out
The withdrawal process has been designed to reduce barriers for users exiting the protocol.
Slush Strategies users can withdraw funds directly through Slush Wallet. Lockups have been removed from the WAL vault, while SUI, WAL, DEEP and USDC strategies can be withdrawn immediately. AlphaFi has also stopped the ALPHA vault and canceled its lockup period.
New deposits and new loans are no longer being accepted. Existing borrowers are instead being encouraged to unwind their positions, while AlphaFi says it will continue operating the withdrawal infrastructure until the final user exits.
The protocol has not announced a deadline for completing the wind-down.
AlphaFi and AlphaLend had previously grown into meaningful parts of Sui’s DeFi market. In Sui’s Q2 2025 DeFi roundup, AlphaLend was listed with $137 million in total value locked and $73 million borrowed at the end of the quarter. AlphaFi itself held $42 million as a yield aggregator.
The lending product allowed users to deposit digital assets, use those assets as collateral and borrow against them. Like other overcollateralized lending protocols, its risk controls depended heavily on accurate asset pricing because the protocol needed to know continuously whether the value of collateral remained sufficient to support outstanding debt.
An ALPHA Oracle Configuration Error Is at the Center of the Investigation
Separate reporting on the incident has attributed the bad debt to an oracle configuration error involving ALPHA. The Sui Foundation’s security team reportedly identified the problem and worked with AlphaFi on the response.
AlphaLend’s architecture uses oracle prices as part of its lending risk system. Borrowing, collateral withdrawal and liquidation depend on current asset valuations, meaning an incorrect price can distort the protocol’s assessment of whether a loan remains adequately collateralized.
However, a detailed technical post-mortem has not yet been published. AlphaFi has not publicly established exactly what was misconfigured, what incorrect ALPHA valuation was accepted, how long that condition persisted, how much borrowing occurred because of it or why liquidation mechanisms did not prevent the positions from developing into bad debt.
Those details matter because an oracle failure can create several different kinds of lending problems. If collateral is valued too highly, borrowers may be able to take out more debt than their collateral should support. If the system subsequently corrects the price, the protocol can discover that the collateral available for liquidation is worth less than the borrowed assets that need to be repaid.
The pattern has parallels elsewhere in DeFi, where apparently healthy collateral can suddenly prove insufficient once an underlying pricing or accounting assumption breaks. The recent Nomic exploit left Osmosis’s allBTC underbacked, forcing the ecosystem to consider using protocol-controlled assets and frozen attacker funds to repair the deficit.
AlphaFi’s situation is different technically, but the economic lesson is similar: solvency can depend as much on whether a protocol measures collateral correctly as on whether its smart contracts execute exactly as written.
Covering the Bad Debt Did Not Save the Protocol
The most interesting part of AlphaFi’s decision comes after the emergency.
If the bad debt is covered, users can withdraw and the protocol is solvent, why shut the business permanently?
The answer may be that recapitalization fixes the balance sheet without fixing the conditions required to operate the product with confidence.
Replacing lost or missing assets is largely an accounting solution. Restarting a lending market requires something harder: confidence that the same risk path cannot produce another deficit.
That means AlphaFi would need to understand the oracle failure, prove that the configuration has been corrected, reassess every market that relies on similar infrastructure, test liquidation behavior and convince depositors that their funds should return after they have just been encouraged to leave.
Other crypto incidents show the same distinction. MultiversX’s recovery from a VM-level exploit required more than fixing the vulnerable code; the network also had to establish that the restored blockchain state could be trusted. Technical remediation and economic recovery are related, but they are not the same thing.
For a lending protocol, confidence is particularly difficult to rebuild because deposits are the raw material of the business. Once users withdraw liquidity, borrowers have less capital available, revenue falls and the economics of continuing to maintain and secure the protocol can deteriorate quickly.
The Oracle Error May Expose a Bigger DeFi Design Weakness
There is also a deeper question about how much damage one incorrectly configured asset should be capable of causing.
Oracle failures are not new. What matters is whether the surrounding risk architecture limits the consequences when one occurs.
A mature lending system can layer defenses around the oracle itself: conservative loan-to-value ratios, borrowing caps, collateral limits, liquidation thresholds, price-deviation checks, stale-price protections and limits on how much exposure can accumulate around a relatively volatile token.
The eventual AlphaFi post-mortem therefore needs to explain more than which oracle setting was wrong. It needs to establish why that error was able to progress into severe undercollateralization before it was detected.
The issue resembles the broader security question raised by the KelpDAO rsETH exploit, where infrastructure behaved according to a configuration that ultimately proved unable to withstand a failure elsewhere in the system. In DeFi, a configuration choice can be just as consequential as a coding vulnerability when large amounts of collateral depend on it.
That may also explain why AlphaFi appears to have concluded that shutting down is preferable to rebuilding.
A protocol can replace bad debt once. Continuing operations means accepting responsibility for the next failure as well.
Solvency and Viability Are Two Different Things
AlphaFi’s shutdown is a useful reminder that being solvent is not the same as having a viable protocol.
Solvency answers one immediate question: can users currently be paid?
Viability asks whether the platform can attract deposits again, generate enough revenue to justify continued development, maintain adequate security and persuade users that its risk model still works.
Those are harder problems to recapitalize away.
Payy recently faced the opposite version of that problem when a $1.83 million bridge exploit forced it to shut down payment functions. The dollar value of an incident can be smaller than the operational consequences when the failure affects infrastructure central to the product.
For AlphaFi, the oracle system sits close to that core. Lending depends on deciding how much collateral is worth, how much users can borrow against it and when a position must be liquidated. If confidence in that calculation disappears, the protocol’s central financial function becomes difficult to restart safely.
The final outcome for users currently looks considerably better than a disorderly insolvency. Withdrawals remain open, lockups have been removed and AlphaFi says the debt hole has been covered.
But that relatively clean exit may be precisely why the shutdown matters.
AlphaFi had enough support to restore solvency and still decided not to continue. The eventual technical report will need to show whether that decision reflected the severity of the oracle failure, the cost of rebuilding its risk architecture, damaged confidence in AlphaLend or a broader judgment that operating the protocol was no longer worth the risk.
Until those details are published, covering the bad debt closes the immediate financial hole. It does not answer the more important question raised by AlphaFi’s exit: what failed badly enough that even a solvent protocol no longer wanted to reopen?
Johan Shamshad is a financial markets writer at Dave Finances covering cryptocurrencies, trading platforms, brokers, fintech, financial regulation, and developments across global markets. He previously worked at Gulf News, adding newsroom experience to his coverage of fast-moving financial and digital-asset markets.
His work focuses on identifying market-moving events, company developments, regulatory changes, product launches, and shifts in trading and financial infrastructure.
Johan contributes news and analysis designed to help readers understand not only what happened, but why a development matters and how it may affect the wider financial landscape.