Raydium exploit: $1.34m stolen from Solana legacy pools
Raydium, a Solana-based DEX, confirmed a Raydium exploit on Wednesday that drained more than $1.34 million from five deprecated liquidity pools tied to a legacy automated market maker (AMM) program.
The incident impacted the Raydium legacy AMM, resulting in stolen tokens including SOL, USDC, and Raydium’s native token RAY. According to a Raydium contributor (0xInfra), no current users could interact with these pools via the Raydium UI because the pools were already deprecated.
The exploiter used a Solana address ending in “Bq33QVk” and reportedly bypassed validation logic in the deprecated program to mint new liquidity provider tokens. The attacker netted nearly $900,000 in USDC, about $357,000 in SOL, and roughly $86,000 in RAY.
Raydium said it will repay the affected funds using its treasury. 0xInfra also stated the issue was not caused by a key compromise or an authority-level problem, and that existing mainnet programs prevent this type of vulnerability.
The Raydium exploit arrives amid a broader wave of DeFi hacks and newly disclosed vulnerabilities. The article notes prior major incidents on Solana (KelpDAO, Drift Protocol) and highlights how frontier AI tools have been accelerating vulnerability discovery, citing a Zcash Orchard pool issue that led to a sharp ZEC drop after an AI-assisted disclosure. While the report says there is no evidence AI was used in this Raydium exploit, the market context may still heighten risk sentiment around DeFi security.
Neutral
This is a security negative in headline terms, but likely limited for traders because the pools were already deprecated and Raydium says it will repay using its treasury. That reduces tail risk of broader loss, while still reinforcing near-term caution around Solana DeFi contracts.
Short term, RAY may face selling pressure as traders price in “DeFi hack” headlines and uncertainty around legacy AMM code paths. SOL and USDC are mentioned as stolen assets, but since the affected pools were not accessible via the UI, contagion to core liquidity and mainstream users should be smaller.
Longer term, the broader takeaway is risk management: recent incidents across crypto (including AI-assisted discovery examples) suggest vulnerabilities may be found faster. Even if this specific Raydium exploit doesn’t involve AI, the market could continue to reward platforms with rapid patching, formal verification, and reduced reliance on legacy logic.
Similar past patterns: after DEX exploits, tokens tied to the platform often trade with elevated volatility until (1) scope is clarified, and (2) repayment/compensation is credible. Here, both points are addressed, which is why the net market impact is best viewed as neutral rather than bullish or deeply bearish.