News linked to both this project and an event.
Odaily News, Spark's strategy lead monetsupply.eth posted on X, stating that as the stablecoin market begins to face a liquidity shortage, the situation is entering a more dangerous phase, in my opinion. Approximately 16.5% of the ETH market is backed by rsETH. If losses on rsETH-backed loans are shared across the mainnet and external chains, they could face a 10% to 15% cut in emode, with the remaining 2% to 3% cut left for ETH suppliers to flatten the umbrella structure. ETH suppliers naturally tend to exit as soon as possible to avoid this risk, so utilization is locked at 100%, and the borrowing rate is insufficient to incentivize the repayment of unrelated LST loops (wstETH, weETH) to release liquidity. Since ETH cannot be withdrawn, users who borrowed stablecoins like USDT using ETH as collateral cannot close their positions even when stablecoin borrowing rates rise, which cuts off the typical incentive mechanisms that maintain market health. Currently, there are two unhealthy incentives causing market utilization to be locked at 100%:1) ETH holders cannot close positions to maintain a healthy LTV, and liquidators cannot atomically withdraw or sell collateral. A drop in the ETHUSD price could lead to bad debt.2) Users supplying USDT, in order to exit their holdings, tend to maximize borrowing of other stablecoins. This position is currently generating positive yield (temporarily), so the exit cost is low; if conditions worsen, they can recover at least 75% of the position's value.The bottom line is that for these pooled/restaking lending markets to function properly, liquidity must be maintained at all costs. The recent weakening of the slope2 for Aave's maximum borrowing rate is having a negative impact and significantly increasing the risk of cascading market failure.
Monetsupply.eth, Strategy Lead of Spark Protocol, posted on X stating that in January this year, low-utilization assets such as rsETH were delisted, and the scope of acceptable collateral and protocol functionalities has been continuously tightened. At the time, this move triggered strong backlash from users employing “ETH circular leverage” strategies. Additionally, Spark has long imposed relatively high maximum interest rate caps on its ETH lending market. Over the past year, Spark has ceded part of its business and revenue to Aave—whose ETH borrowing rates at one point dropped to 10% or lower. However, amid the current market crisis, this strategy has proven more prudent: SparkLend still maintains ample ETH withdrawal liquidity, whereas Aave is experiencing liquidity strain—or even “locking”—across Ethereum mainnet and multiple Layer-2 chains including Arbitrum and Base. Monetsupply.eth further warned that, since ETH serves as the core collateral asset, when market utilization reaches 100%, liquidations of collateral will fail to execute normally. Liquidity exhaustion not only degrades depositors’ experience but may also pose systemic risk. Given Aave’s current liquidity shortage, a 15–20% drop in ETH’s price could trigger significant bad debt accumulation—exacerbated by potential fallout from the rsETH incident.
According to the South China Morning Post, Interconnects AI, a U.S.-based AI tracking firm, released a report stating that as of March 2026, Alibaba Cloud’s Qwen series models accounted for over 50% of global open-source model downloads, with a cumulative total of 942.1 million downloads—far surpassing competitors such as Meta’s Llama and DeepSeek. In February alone, Qwen downloads reached 153.6 million—exceeding the combined total downloads of the next eight major vendors. The report notes that Qwen’s dominant position stems from the exceptional popularity of its smaller-parameter variants (under 10 billion parameters), which enable developers to customize and deploy models freely at low cost. Since the launch of Qwen 2.5 in September 2024, Chinese models have begun outpacing mainstream U.S. open-source models like Llama; the release of Qwen 3.5 in February this year further solidified its lead. Meanwhile, open-source strategy has become a critical battleground in the U.S.-China AI competition. Meta has abandoned its open-source approach this year, instead launching the closed-source flagship model Muse Spark. Similarly, Chinese vendors including Alibaba Cloud and Zhipu AI have shifted some of their latest models to closed-source to expand direct commercialization channels.