Aave Yields Fail to Reflect Risk as DeFi Losses Mount, Ledn Exec Says

The debate has intensified following turmoil at Aave – a DeFi lending protocol on which users deposit assets to earn yield or borrow against collateral. Since mid-April, the protocol has been grappling with the fallout from an external exploit that saddled the platform with an estimated $200mn in potential bad debt. 

The incident sparked a rapid withdrawal of funds in classic bank run style, with roughly $18bn leaving Aave and total value locked dropping by about a third. As liquidity dried up, borrowing demand surged, pushing stablecoin rates sharply higher. 

Borrowing costs for stablecoins USDC and USDT on Aave jumped from around 3.4% to as high as 14–15% in the wake of the exploit, as users scrambled to access liquidity while lenders pulled funds from pools. 

Stress on pools

Aave uses a dynamic interest rate model designed to manage liquidity by incentivizing or discouraging borrowing based on the "Utilization Rate." The system operates on a dynamic curve that keeps borrowing costs low when capital is plentiful but triggers a sharp interest rate spike once utilization hits the "optimal" threshold. This pricing mechanism acts as a circuit breaker, forcing borrowers to repay loans while attracting new liquidity providers with high yields. 

But the supply-demand dynamics that set rates do not fully account for the broader risk profile, according to Ledn co-founder Mauricio Di Bartolomeo. Ledn is a crypto lender that offers Bitcoin-backed loans at fixed rates starting near 10%, significantly above pre-incident yields on Aave. 

"Aave depositors were getting paid 2.32% before the hack. That is less than the Federal Reserve. That makes no sense," Di Bartolomeo told Sandmark in an interview at the Bitcoin 2026 conference.  

He argued that returns in DeFi lending markets appear disconnected from the actual risk profile, pointing to what he described as "mispriced" incentives driven by excess liquidity and complex leverage strategies. 

Risk not priced 

Di Bartolomeo said the structure of protocols like Aave makes it difficult to assess risk consistently, particularly in stress scenarios. Unlike traditional finance, where participants and exposures are known, DeFi pools operate with pseudonymous users and have limited transparency over counterparties. 

"Nobody knows who's in the pool… where the money is coming from or what it's being used for," he said. 

A failure to reprice yields after losses would suggest other factors – such as anonymity – are driving demand.  

Uneven losses 

He also pointed to dynamics embedded in DeFi lending models, in which outcomes depend on timing rather than proportional risk-sharing. "If something bad happens, everybody can get out. The people that got out first, their loss is zero. The people that got out last, their loss is everything," he said. 

That uneven distribution of losses complicates risk modelling, particularly for institutional investors, who typically require predictable exposure profiles. "You might estimate a 2% loss, but one person gets zero and another gets 50%," he said. 

Institutional limits  

Di Bartolomeo said such dynamics make DeFi structurally difficult for institutions to access at scale, despite expectations that banks could increasingly connect to onchain lending markets.  

He added that while decentralized and regulated systems can coexist, they serve different use cases, with DeFi prioritizing accessibility and privacy, often at the expense of safeguards common in traditional finance.  

"My thesis is that interest rates typically reflect the risk quality of an investment. And if we are pricing risk, and we have demonstrated that there is risk in DeFi, in every DeFi protocol, it should price higher." This repricing of risk, according to Di Bartolomeo, will be a key test for the sector, particularly whether yields adjust to reflect the potential for loss.