Jupiter Lend Clarifies Earlier 'Zero Contagion' Statement Was Not Accurate

Jupiter Exchange is working to clear the air after concerns emerged around its lending protocol. Kash Dhanda, the platform’s chief operating officer, acknowledged that earlier communications overstated the safety of Jupiter Lend’s vaults. Previous posts had suggested there was zero risk of contagion, a claim that was later removed. Dhanda admitted this messaging missed the mark and stressed the need to provide clearer, more accurate information about the risks involved.

Jupiter Clarifies Vault Risks and Isolation

Before Dhanda issued his clarification, Jupiter had shared social posts describing the vaults as operating with “isolated risk,” even suggesting that this setup prevented any interaction between pairs. One post went further, claiming the design eliminated the possibility of contagion entirely. The message drew criticism and was eventually removed, prompting a broader review of how the protocol’s mechanics were being communicated.

In a video lasting a little over three minutes, Dhanda acknowledged the error in the video and stated that the posts had been removed to prevent the incorrect message from spreading, adding that the vaults do carry a limited level of contagion risk, rather than none.

In hindsight, we should have issued a correction right when we deleted it. But so it goes. The point is, what we should have said is that there is a very limited risk of contagion because that is actually accurate.

Kash Dhanda

Fluid co-founder Samyak Jain added context, explaining that Jupiter Lend employs rehypothecation. He noted that the vaults are still isolated, stating that “each vault has its own configs, caps, liquidation threshold, liquidation penalty, etc and there’s surely rehypo for better capital optimization.”

Jupiter Lend Faces Criticism Over Vault Structure

However, Kamino co-founder Marius Ciubotariu voiced concern that this setup could mislead users. He stated that the Jupiter Lend migration tool had been blocked because their “users have been misled about the protocol design and the risks they are taking.”  Ciubotariu argued that while the platform had long indicated the vaults would prevent one asset from impacting another, the system operates differently in practice. 

The Kamino co-founder explained that when a user supplies SOL and borrows USDC, the deposited SOL can be lent to loopers like JupSOL and INF, exposing the user to the risks associated with those positions. In his view, this results in fully interconnected risk across assets, which contradicts the way the product has been presented.

Dhanda, however, maintained in the video that the protocol does use rehypothecation but still keeps the vaults isolated. He said the structure should be seen as isolated because of how each vault is configured.

Resistance and Support in Protocol Design

The word “isolation” appeared to be at the center of disagreement between Jupiter Lend executives and Marius Ciubotariu:

• Dhanda and Jain maintain that the vaults are isolated because each operates with its own configuration, including loan-to-value ratios, limits on assets, and rules for liquidations, which provide resistance and allow liquidity to flow through rehypothecation
• Ciubotariu countered that this setup falls short of true isolation, arguing that in both traditional and decentralized finance, users need to know if collateral can be reused and how that affects contagion risk
• He emphasized that describing pair-specific configurations as isolated is misleading and could give users a false sense of security

Meanwhile, in comments shared with The Block, Ciubotariu noted that he would consider reopening the migration tool once Jupiter stops presenting the system in a way that gives users and the wider Solana community an inaccurate picture, as well as updates the tool to allow movement in both directions.

Early Performance and Safety Debate

Jupiter Lend was introduced in August with loan-to-value ratios reaching up to 90%. Dhanda pointed out the protocol’s performance during the market drop on October 10, noting that its ability to avoid any bad debt demonstrated it could handle pressure even as a relatively new platform.

However, Ciubotariu disagreed, pointing out that the protocol had only been active for a month with limited positions exposed to real market stress. He noted that Jupiter Lend would need years of testing under tougher conditions before making claims about safety or resilience.