Public blockchains leak almost everything. Orderflow, wallet history, intent, slippage patterns, and execution timing.

On high-throughput chains like Solana, this transparency has created a measurable MEV crisis. Sandwich attacks now affect ~2.9% of trades and have extracted over $2B cumulatively (64k SOL in the last 6 months).

This is part of why privacy infrastructure is becoming one of the fastest-growing categories in crypto (up +251% YTD), with TVL recently hitting new all-time highs. Yet most solutions focus on user privacy, not execution privacy, where the actual value leakage happens.

DarkVeil takes a different approach. It uses encrypted orderflow, secure enclaves, and an AI-driven execution engine to match trades privately at oracle-verified prices, eliminating slippage and blocking MEV vectors like frontrunning and sandwich attacks. Only the final settlement hits the chain, not the orderflow itself.

The aim is to recreate the “dark pool” model from traditional markets, but with deterministic execution and non-custodial guarantees.

Would love feedback from the HN community on: - Feasibility of enclave-based matching on high-throughput chains - Tradeoffs of private vs public liquidity - AI-assisted execution models vs traditional AMMs