Key Takeaways:
Institutional investors are largely shunning decentralized exchanges offering perpetual futures, held back by persistent security risks and fundamental compliance gaps, according to a panel of experts at Consensus Miami.
"I’m scared to use DeFi right now," Michael Anderson of Canary Labs said. "It does feel like a bit of a minefield, and you’re just waiting for the next headline each day."
The discussion highlighted recurring exploits within decentralized finance, with panelists pointing to the recent multi-million dollar hack of Solana-based DEX Drift Protocol as a prime example of risks that give institutions pause. The April exploit was one of several high-profile security incidents this year, a period that also saw Kelp DAO move its restaking token to Chainlink's oracle network after an exploit it attributed to LayerZero's infrastructure.
For perpetual DEXs, the inability to attract institutional-grade liquidity represents a major growth ceiling. Veteran trader Wizard of SoHo said the next major competitive battleground for these platforms will be whether any of them can solve the puzzle of safely onboarding institutional capital.
A primary challenge is the structural conflict between DeFi's open, permissionless ethos and the strict know-your-customer (KYC) and compliance obligations governing institutional finance. Anderson noted that while some activity has moved to DEXs from Asia amid tighter KYC enforcement on centralized exchanges, the model remains a difficult fit for larger, regulated firms.
"Crypto wants to be more non-KYC," Anderson said, "but to bring on institutional [players] you need to have some form of KYC at the larger size." He argued it will be "very difficult" for large firms to use decentralized platforms at scale compared with their centralized counterparts.
The panel also touched on the rise of AI-driven trading. MN Fund founder Michaël van de Poppe characterized AI agents as the next evolution of algorithmic trading, predicting a future where manual trading becomes obsolete.
"We are not going to be trading ourselves anymore. Nothing will be manual," van de Poppe said. "AI agents will be doing it for us, and they are probably better."
He cautioned, however, that the effectiveness of these tools depends entirely on the quality of their programming, a concern that echoes broader industry warnings. Security researchers recently found thousands of web apps built with AI coding tools that exposed sensitive corporate and personal data due to a lack of basic security configurations, highlighting the risks of rapid, unchecked innovation.
This article is for informational purposes only and does not constitute investment advice.
