Crypto’s Privacy Tech Wave

With growing concerns around surveillance and data exploitation, the crypto space has recently accelerated efforts to integrate privacy-enhancing technologies (PETs) into its core infrastructure.

Blockchains are radically transparent by design and while the crypto industry has long valued privacy methods — coin mixers or privacy-based tokens — it has also struggled to expand what can be private – beyond simple DeFi and payments – without siloing that privacy to specialized networks.

With blockchains increasingly used for AI training and institutional finance, applications using alternative cryptographic techniques are becoming popular. Four technologies in particular are catching steam: Multi-Party Computation (MPC), Fully Homomorphic Encryption (FHE), Trusted Execution Environments (TEEs), and Zero-Knowledge Transport Security Layer (zkTLS).

This article aims to showcase each technology's role in enhancing privacy, use cases, and key projects building on each. 👇

Multi-Party Computation (MPC)

MPC is distributed computation that allows multiple groups to collectively compute something without revealing their own information.

Say that you and five friends want to find your average salary without revealing individual amounts. Each person splits their salary into six random shares and sends one to each person. Everyone holds one share from each person, but no one can reconstruct anyone's salary because they only have one of six needed shares. Everyone performs math on the shares — not original salaries. These results combine to compute the final average without anyone learning individual salaries.

MPC is particularly valuable when regulatory constraints or competitive concerns prevent direct data sharing, but collective analysis would benefit all parties. Multiple hospitals wanting to train AI on patient data is a classic example — regulation prohibits sharing sensitive medical data, but MPC enables collective training without actually sharing data.

Roadblocks for MPC

As more people join an MPC network, management becomes harder. The system needs more messages between participants, slowing things down due to internet capacity limits. Each person does more calculations, using more computing power. While blockchains can discourage cheating by punishing bad actors of the network who may want to collude, they don't solve these resource and computing power problems.

Who's Using MPC and for What?

• Fireblocks — Institutional custody using MPC to split private keys across devices so full keys are never exposed.
• Arcium — Chain-agnostic network using MPC for private AI processing and sensitive tasks.
• Renegade — Onchain dark pool using MPC for confidential trading.

Demystifying Crypto ‘Dark Pools’ on Bankless

Will dark pools make crypto trading safer or just sideline retail investors?

BanklessDavid C Jun 3, 2025 • 6 min read

Fully Homomorphic Encryption (FHE)

FHE allows data processing without decryption, meaning sensitive data remains encrypted when stored, transmitted, and analyzed.

Currently, data is encrypted during transmission but must be decrypted for processing, creating vulnerability windows. For example, when I send photos to the cloud, they're encrypted in transit but decrypted upon arrival. FHE eliminates this decryption step — data stays encrypted throughout computation, protecting information during active use.

Think of FHE like a locked safe with programmable gloves. You put private data inside and program instructions: "add these numbers," "sort this list." You send the safe and gloves to someone else. They operate on contents blindly, following instructions without seeing what's inside. When finished, they return the safe, and you unlock it to find the correct result.

Roadblocks for FHE

FHE comes with major performance penalties — computations are 10-100x slower. Adding ZK verification (zkFHE) makes it even slower by several orders of magnitude. Developers want this combo because while FHE protects input, it doesn't guarantee correct operations. In other words, it's the problem of whether someone you authorized to run a computation on the FHE-protected data actually did so correctly. This verifiability is missing, but adding it makes an already slow system nearly unusable for real-time applications.

Who's Using FHE and for What?

• Zama — FHE tooling provider enabling encrypted smart contracts on EVM networks with a fhEVMs, among other tools.
• Fhenix — Research company bringing FHE to real-world applications.
• PrivaSea — AI training network using Zama’s FHE tooling for encrypted machine learning.
• Octra — Universal chain using proprietary FHE for high-speed encrypted computation with machine-learning consensus and rentable services.

Trusted Execution Environments (TEEs)

TEEs are secure hardware zones that store and process data in isolation, preventing the rest of the machine — including the OS and operator — from accessing that data.

If you have an iPhone, you interact with TEEs daily since Apple uses them for biometric data. They work like this: TEEs store face or fingerprint scans inside secure chip zones. When apps request authentication, new scans are sent into the TEE for comparison. This matching happens inside sealed hardware — no biometric data is visible to apps or OS. The TEE simply returns yes or no.

TEEs have begun appearing in crypto for confidential smart contracts and computing. Unichain, Uniswap's L2, uses TEEs to build blocks fairly and prevent MEV attacks.

Roadblocks for TEEs

TEE integrity relies on hardware vendors, not distributed networks, making them centralized by crypto standards. Someone could compromise TEEs in production or exploit weaknesses. This happened to Secret Network when researchers found Intel chip weaknesses that decrypted all network transactions.

Who's Using TEEs and for What?

• Space Computer — Blockchain using TEEs on satellite nodes, making hardware tamper-proof by running in orbit.
• Oasis Protocol — Layer 1 using TEEs for confidential smart contracts with EVM-compatible runtimes.
• Phala Network — Decentralized cloud platform using TEEs from multiple hardware providers for confidential computing.

zkTLS

zkTLS merges TLS (already used in HTTPS for internet security) with zero-knowledge proofs (ZKPs) to keep information private yet verifiable.

By adding ZKPs, zkTLS allows users to transmit any HTTPS data (95% of web traffic) while controlling the information revealed. This lets any Web2 platform data function as a public API regardless of platform permissions, connecting the entire web and bridging Web2 and Web3.

For example, say you want to use your bank balance for an onchain loan. You access your bank via a zkTLS tool, which can analyze any displayed data since banks use HTTPS. The tool generates a ZKP of your balance — proving funds without revealing exact amounts or transaction history. You submit this proof to DeFi lending platforms, which verify creditworthiness without accessing private financial data.

Roadblocks for zkTLS

zkTLS only works with data that websites already display — it can't force sites to reveal hidden information. It depends on continued TLS protocol use and requires real-time oracle involvement, introducing latency and trust assumptions.

Who's Using zkTLS and for What?

• ZKP2P: On/off ramp protocol using zkTLS to privately move funds on and off chain.
• EarniFi — Lending platform using zkTLS for privacy-preserving loans to employees with earned but unpaid wages.
• DaisyPay — App using zkTLS for influencer collaboration and immediate payouts.

Onramp Privately with ZKP2P on Bankless

Explore how ZKP2P’s V2 offers a privacy-preserving, low-cost way to move funds on and offchain

BanklessDavid C Jun 20, 2025 • 5 min read

Overall, each PET serves different goals with distinct trade-offs. Applications will likely combine multiple PETs depending on data needs. A decentralized AI platform might use MPC for initial coordination, FHE for computation, and TEEs for key management. 

There are many different approaches to zkTLS which utilize one or another of the other PETs in their architecture. Together, these tools can vastly expand crypto's design space and realize its potential as the next web iteration. As we all know, crypto still needs to work on its user experience, which will be particularly critical to making these privacy services more usable and widely adopted.