Privacy Tech: TEE, FHE & MPC

The Privacy Problem in Blockchain

Public blockchains are transparent by design—every transaction is visible to everyone. While great for trustlessness, this creates real problems:

• Front-running — Bots see your trades and extract value
• Competitive intelligence — Rivals can track your on-chain activity
• Compliance barriers — Enterprises can't expose sensitive data
• User privacy — Your entire financial history is public

This is where confidential computing comes in—technologies that enable computation on private data while keeping that data hidden.

The Three Main Approaches

Three technologies dominate confidential computing for blockchain. Each takes a fundamentally different approach:

TEE: Trusted Execution Environments

TEEs create isolated "enclaves" within the CPU where sensitive code and data are protected from the rest of the system—including the operating system, hypervisor, and other applications.

How TEEs Work

1. Encrypted code and data loaded into isolated memory
2. CPU executes within the enclave using special instructions
3. Memory access controls prevent external reads
4. Attestation proves code is running unmodified in genuine hardware

TEE Implementations

• Intel SGX — Most common, but facing vulnerabilities
• AMD SEV — VM-level isolation
• ARM TrustZone — Mobile-focused
• AWS Nitro Enclaves — Cloud-native option

Blockchain Applications

• Dark pools — Hidden order books for private trading
• MEV protection — Block builders that hide transaction details
• Confidential AI — Running models on private data
• Key management — Secure custody solutions

FHE: Fully Homomorphic Encryption

FHE is often called the "holy grail" of encryption. It enables computations directly on encrypted data, producing encrypted results that decrypt to the correct answer.

How FHE Works

1. Data is encrypted with FHE scheme (lattice-based cryptography)
2. Computations performed on ciphertext (addition, multiplication)
3. Results remain encrypted throughout
4. Only the key holder can decrypt the final result

FHE Progress

FHE was long considered impractical, but recent advances are changing this:

• GPU/FPGA acceleration reducing compute overhead
• New schemes (TFHE, CKKS) more efficient for specific operations
• Hybrid approaches combining FHE with TEE for performance

Research shows FH-TEE (combining both) can be 10x faster than pure FHE while maintaining stronger security guarantees.

MPC: Multi-Party Computation

MPC distributes computation across multiple parties so that no single party ever sees the complete data. Each party holds a "share" of the secret and collaborates to compute results.

How MPC Works

1. Secret is split into shares (e.g., using Shamir's Secret Sharing)
2. Each party holds one share—useless alone
3. Parties interact through cryptographic protocols
4. Final result reconstructed only when needed

MPC Applications

• MPC Wallets — Fireblocks, ZenGo split keys across parties. No single device holds the full private key.
• Threshold signatures — k-of-n signing schemes for institutional custody
• Private auctions — Bid without revealing your price
• Cross-chain bridges — Distributed key management

Head-to-Head Comparison

How Does ZK Fit In?

Zero-Knowledge Proofs (ZKPs) are often mentioned alongside privacy tech, but they serve a different purpose:

ZK proves statements; TEE/FHE/MPC enables private computation. They're often combined—MPC can use ZK proofs for verification, and FHE results can be proven correct with ZK.

Privacy Pools & Regulatory Precedent

The intersection of blockchain privacy and regulation came to a head with the Tornado Cash sanctions — a watershed moment that reshaped how the industry thinks about privacy.

Tornado Cash: The OFAC Precedent

In August 2022, the U.S. Treasury's Office of Foreign Assets Control (OFAC) sanctioned Tornado Cash, a non-custodial Ethereum mixer. This marked the first time a government sanctioned open-source smart contract code rather than a person or organization. Key consequences:

• USDC blacklisting — Circle froze ~$75K in USDC held in Tornado Cash addresses
• Developer arrest — Tornado Cash developer Alexey Pertsev was arrested in the Netherlands and convicted under anti-money laundering laws
• RPC censorship — Infura and Alchemy began blocking transactions to Tornado Cash contracts
• GitHub removal — Tornado Cash repositories were initially removed from GitHub

The sanctions forced the industry to confront a fundamental question: can privacy and regulatory compliance coexist on public blockchains?

Privacy Pools: The Compliant Privacy Model

Vitalik Buterin, along with researchers, proposed Privacy Pools as a potential solution. The concept uses zero-knowledge proofs to create privacy with selective disclosure:

1. Users deposit funds into a privacy pool (like a mixer)
2. When withdrawing, users generate a ZK proof showing their deposit is in an association set — a set of "clean" deposits
3. The proof demonstrates the funds did not come from a sanctioned or illicit source, without revealing which specific deposit is theirs
4. Compliance is achieved without sacrificing individual privacy

Privacy-Preserving Blockchain Projects

Investment Implications

Privacy tech is becoming critical infrastructure. Key trends for investors:

Growing Enterprise Demand

Enterprises need privacy for blockchain adoption. Financial institutions can't expose trade data. Healthcare can't put patient data on-chain without privacy guarantees. This creates real demand for production-ready solutions.

Infrastructure Plays

• FHE-focused chains (Fhenix, Inco Network) building privacy-first infrastructure
• MPC wallet providers (Fireblocks, ZenGo) already capturing institutional custody
• TEE infrastructure (Secret Network, Oasis) live in production

Convergence with AI

AI on private data is a massive use case. Training models on sensitive data (healthcare, finance) without exposing it. FHE and TEE are essential for "confidential AI" which is emerging as a major narrative.

Regulatory Tailwinds

Privacy regulations (GDPR, etc.) push enterprises toward privacy-preserving solutions. Blockchain privacy tech that maintains compliance while enabling on-chain activity has strong product-market fit.

Key Takeaways

1. TEEs provide fast, hardware-based privacy but require trusting chip manufacturers and have known side-channel vulnerabilities. Best for performance-critical applications with moderate trust requirements.
2. FHE is the "holy grail"—pure cryptographic privacy with quantum resistance—but remains computationally expensive. Improving rapidly with new optimizations.
3. MPC distributes trust across parties, eliminating single points of failure. Already production-ready for wallets and custody, but has coordination overhead.
4. Hybrid approaches combine strengths. TEE + FHE can be 10x faster than pure FHE while maintaining stronger guarantees.
5. ZK proves; TEE/FHE/MPC computes. They solve different problems but are often combined.
6. Enterprise adoption is driving demand. Privacy-preserving computation is becoming essential infrastructure for institutional blockchain use.