DeFi Hacked Again for $292 Million, Is Even Aave No Longer Safe?

marsbitPublicado a 2026-04-18Actualizado a 2026-04-18

Resumen

On April 19, a major DeFi security breach occurred, resulting in the loss of approximately $292 million. The attack targeted Kelp DAO’s rsETH bridge contract built on LayerZero, with 116,500 rsETH stolen. The attacker initiated the exploit using funds from Tornado Cash and manipulated the LayerZero EndpointV2 contract to transfer the assets. Kelp DAO confirmed the incident and temporarily paused rsETH contracts across multiple networks while collaborating with security experts for investigation. Initial analysis suggests the root cause was a compromised private key on the source chain, with the contract secured by only a 1/1 validator set, making it vulnerable to a single malicious transaction. The attacker used the stolen rsETH as collateral on lending platforms—including Aave, Compound, and Euler—to borrow more liquid assets like WETH, accumulating over $236 million in debt. Aave alone accounted for $196 million of this amount. In response, Aave froze its rsETH markets and stated it would explore covering potential bad debt through its Umbrella safety module, which holds around $50 million in WETH. This incident follows another large exploit earlier in April, where Drift Protocol on Solana lost $280 million. The repeated high-value attacks raise concerns about DeFi security, even affecting major protocols like Aave. Users are advised to exercise caution, diversify holdings, and limit exposure to on-chain protocols until more robust security measures are established.

Original | Odaily Planet Daily (@OdailyChina)

Author | Azuma (@azuma_eth)

On April 19th, Beijing time, DeFi security suffered another major blow.

On-chain data shows that around 1:35 this morning, the rsETH bridge contract of Kelp DAO, the second-largest liquid staking protocol, based on LayerZero, was suspected to be exploited by hackers, resulting in a loss of 116,500 rsETH, valued at approximately $292 million.

Further tracing the on-chain records, the attacker's address received 1 ETH in initial funds from the mixing protocol Tornado Cash about 10 hours before the incident. Subsequently, this address called the lzReceive function on the LayerZero EndpointV2 contract. This call triggered Kelp's bridge contract, transferring 116,500 rsETH to another attacker address.

Approximately 2.5 hours after the incident, Kelp DAO officially confirmed the attack on X: "Earlier today, we detected suspicious cross-chain activity involving rsETH. During the investigation, we have suspended the rsETH contracts on the mainnet and multiple Layer 2s. Our auditors are working with security experts from LayerZero and Unichain to closely monitor the situation. We will keep you updated on the latest developments. Please follow official channels."

After the incident, various DeFi projects and security agencies analyzed the cause. An analysis by D2 Finance was frequently cited within the community — LayerZero Scan marked the source's counterpart as Kelp DAO, meaning the message came from a legitimately deployed counterpart contract by Kelp itself, and this path had previously recorded 308 message nonces. Therefore, the root cause of this attack is a 'compromise of the source chain private key.'

Steven Enamakel, a developer at TinyHumans AI, added that the contract was secured by only a 1/1 validator set (DVN), meaning a single erroneous transaction from the validator was sufficient to cause the issue.

Hacker Escapes via Aave, Suspected Bad Debt Incurred

Due to the limited trading liquidity of rsETH itself, the hacker's chosen escape strategy was to route through lending protocols like Aave, using rsETH as collateral to borrow more liquid wETH.

Monitoring by PeckShield Alert showed that as of 4:30 this morning, the hacker's address had deposited the stolen rsETH into lending protocols including Aave V3, Compound V3, and Euler, borrowing a large amount of WETH, with a total debt exceeding $236 million — of which Aave alone accounted for $196 million, Compound $39.4 million, and Euler only $840,000.

Following the incident, Aave promptly froze the rsETH market on Aave V3 and V4. The team subsequently issued an official statement on X: "Aave's contracts were not attacked; this attack is related to rsETH. Freezing rsETH is to prevent new rsETH deposits and collateral borrowing while the situation is assessed. We are reviewing the rsETH borrowing information on Aave that occurred after the attack and will share more details as soon as possible."

Shortly after the initial statement, Aave updated the post, adding: "If the protocol accumulates bad debt due to this incident, we will explore avenues to cover the deficit."

As of writing, the specific amount of bad debt caused by this incident is still unclear.

monetsupply.eth, Head of Strategy at Aave's direct competitor Spark, stated that if rsETH experiences a 19% discount (the stolen amount represents 19% of the total rsETH supply), Aave could potentially incur over $100 million in bad debt due to highly leveraged recursive borrowing.

However, Marc Zeller, founder of the representative Aave governance team Aave Chan Initiative (ACI) (who has announced he will leave Aave in July due to governance disagreements), offered a different perspective. Zeller initially advised users to quickly withdraw WETH from Aave V3 to avoid losses and confirmed that the USDC and USDT markets on Aave were unaffected. In response to another user's speculation that 'bad debt could reach hundreds of millions,' he stated: 'Far less than that figure.'

But Marc Zeller also mentioned that it was time to test Umbrella in a real production environment. Umbrella refers to Aave's automatic security module, essentially a pool of funds to handle bad debt. Users can deposit assets into it for higher incentives, but the pool also bears potential losses if the protocol incurs bad debt.

Aave protocol data shows that Umbrella currently holds approximately $50 million worth of WETH that could be used to address potential bad debt from this incident, but it is uncertain whether this will be sufficient to cover the shortfall.

Affected by this event, AAVE's price fell sharply by nearly 10%, trading at around 104.6 USDT at the time of writing.

Another Hundred-Million-Dollar Security Incident in April

This is not the first major security incident this month.

As early as April 1st, the Solana生态衍生品交易协议 Drift Protocol was attacked, losing up to $280 million (see 《April Fool's Joke? Drift Protocol Hacked for Over $280 Million, Possibly Becoming Solana's Second-Largest DeFi Heist》).

Afterwards, Drift Protocol directly blamed the hack on "North Korean hackers," but fortunately, institutions like Tether pledged $147.5 million for user compensation, giving users some hope for reimbursement.

Just over ten days later, another, larger hack occurred. How will this one be resolved?

Is There Any Safe Place Left in DeFi?

Security issues in DeFi are intensifying.

On one hand, there are continuous hacking incidents; on the other, there are persistent security threats posed by AI like Mythos (refer to 《Odaily Interview with Yu Xian: How Does the Leak of Anthropic's Nuclear-Grade New Model Affect Crypto Security Offense and Defense?》). For DeFi users, the previous countermeasure was to concentrate funds towards well-audited, reputable top-tier protocols. But now, even top-tier protocols like Aave, which retail users subconsciously considered extremely unlikely to have problems, are indirectly affected. Where can users move their funds?

Personally, it is currently not advisable for users to keep large amounts of funds on-chain. If there is a genuine need, please ensure proper diversification and isolation of positions.

As of writing, many details regarding this incident remain unclear. Odaily will continue to follow the developments. Please stay tuned.

Preguntas relacionadas

QWhat was the total value of rsETH stolen in the Kelp DAO attack?

AThe attack resulted in the loss of 116,500 rsETH, valued at approximately $292 million.

QWhich lending protocol did the hacker use to borrow WETH using the stolen rsETH as collateral?

AThe hacker used Aave V3, Compound V3, and Euler to borrow WETH, with Aave V3 accounting for the largest debt of $196 million.

QWhat was the suspected root cause of the Kelp DAO bridge contract exploit according to D2 Finance's analysis?

AThe root cause was identified as a compromise of the source chain private key, allowing the attacker to send a malicious message from a legitimate Kelp DAO endpoint contract.

QWhat mechanism does Aave have to cover potential bad debt from this incident, and how much capital is currently available in it?

AAave has an automatic security module called Umbrella, which currently holds about $50 million in WETH to cover potential bad debt, though it's uncertain if this will be sufficient.

QHow did Aave respond immediately after the attack was discovered?

AAave froze the rsETH markets on Aave V3 and V4 to prevent new deposits and collateralized borrowing, and announced they were assessing the situation and exploring ways to cover any resulting bad debt.

Lecturas Relacionadas

From South Korea to the United States: Blue-Collar Jobs Are Becoming Increasingly Popular, Thanks to AI

AI is reshaping the labor market's value proposition. The traditional four-year college degree is losing its appeal as a guaranteed career path, while skilled blue-collar trades like electricians, welders, and plumbers are experiencing historic demand and wage premiums. This shift is driven by dual pressures: AI's displacement of certain white-collar roles and a booming need for physical infrastructure and data center construction. Data confirms the trend. In the U.S., vocational school revenue surged, and a significant portion of recent layoffs are AI-related. Surveys show a majority of Gen Z adults plan to pursue blue-collar work, citing better job security against AI automation. Vocational education interest has exploded recently. Experts cite a psychological shift as younger generations seek tangible, AI-resistant careers and avoid high student debt. In many cases, salaries for skilled trades now match or exceed those requiring a bachelor's degree. In South Korea, semiconductor vocational high schools boast near-total employment, with graduates securing high-paying roles at companies like Samsung. The shortage is structural, exacerbated by a retiring baby boomer workforce and massive infrastructure projects. Companies like JPMorgan Chase, Meta, and Lowe's are investing heavily in training programs. However, overcoming historical stigma and a "perception gap" around trade careers remains a key challenge to closing the talent gap.

marsbitHace 2 min(s)

From South Korea to the United States: Blue-Collar Jobs Are Becoming Increasingly Popular, Thanks to AI

marsbitHace 2 min(s)

Qualcomm: AI Hype Subsides, When Will Smartphones Emerge from the Gloom?

Qualcomm reported its Q3 FY2026 results (ending June 2026), with revenue of $9.95B, down 4% YoY but above expectations. Gross margin declined to 53.1%, pressured by rising costs across manufacturing and memory. Key business segments showed mixed performance: Handset revenue fell 19.6% YoY to $5.09B, dragged by an 11% decline in non-Apple Android shipments and weaker high-end mix. Conversely, Automotive revenue surged 61% to $1.59B, and IoT grew 9% to $1.83B. Core operating profit dropped 41% YoY due to margin compression and higher expenses. Management's Q4 FY2026 guidance projects revenue of $9.7B-$10.5B, in line with consensus, but Non-GAAP EPS guidance of $2.05-$2.25 fell short of expectations. Amidst persistent weakness in its core handset market, Qualcomm is pursuing growth in AI, focusing on Edge AI (smartphones, PCs, automotive) and Data Center AI. Its data center strategy includes four pillars: AI accelerators (e.g., AI200), commercial CPUs (Dragonfly C1000), custom silicon, and connectivity solutions. While these initiatives initially boosted its stock, concerns over AI capital expenditure sustainability have since erased those gains. The company targets $5B in data center revenue for FY2027 and $15B for FY2029. The report concludes that with the traditional handset business still under pressure, the data center opportunity is currently viewed as a longer-term option, and a more conservative valuation based on core operations may be warranted until AI contributions materialize.

marsbitHace 7 min(s)

Qualcomm: AI Hype Subsides, When Will Smartphones Emerge from the Gloom?

marsbitHace 7 min(s)

From TPU to Self-Evolving Agents: How Jeff Dean Predicts the Next Step in AI

At the 2026 YC Startup School, Jeff Dean outlined his vision for AI's next phase, shifting focus from simply scaling models to building intelligent, autonomous systems. He believes AI's progress is no longer just about creating smarter models, but about integrating them into systems capable of long-term, iterative work, automated experimentation, and continuous learning. This evolution moves the competition from "who has the bigger model" to "who can best organize intelligence." Dean suggests AI capabilities are now comparable to a junior engineer, enabling the automation of complex workflows. However, the true challenge and opportunity lie in managing these AI "workers" at scale. He emphasizes the importance of **context engineering**—structuring tools, memory, and feedback loops—over raw model power. For startups, this means building deep expertise in niche domains where general models currently fail (near 0-1% success rates), leveraging proprietary data, specialized tools, and domain-specific evaluators. A recurring theme is re-examining fundamental constraints. Dean's past work, like moving Google's search index to memory or creating the TPU, stemmed from questioning outdated assumptions about hardware and cost. He sees similar inflection points today, particularly in **specialized inference hardware** to drastically reduce latency and energy consumption for real-time Agent operation. Notably, he points out that in modern AI systems, the dominant cost is often not computation but **data movement**. Reliable, long-running Agents require robust system design, borrowing concepts from distributed computing like checkpointing, state management, and parallel exploration to handle failures and maintain progress over days or weeks. As AI automates execution, the scarcest human skills will shift to **defining clear specifications**, **judging what problems are worth solving** (taste), and designing effective feedback loops. Ultimately, Dean's framework prioritizes understanding the problem deeply, identifying the true bottlenecks, and systematically building closed-loop systems where AI can not only perform tasks but also improve AI itself.

marsbitHace 7 min(s)

From TPU to Self-Evolving Agents: How Jeff Dean Predicts the Next Step in AI

marsbitHace 7 min(s)

Trading

Spot
活动图片