Privacy Ecosystem on Solana
This article is an excerpt from our research on the Full-Stack Privacy Ecosystem.
Download the complete report here:
https://docsend.com/v/sjv2g/thefullstackprivacyecosystem
The privacy ecosystem on Solana is nascent.
We had the opportunity to chat with Mert and ask him a couple of questions regarding privacy on Solana. In his own words, he said Solana is lagging a bit on privacy.
But what does a mature ecosystem look like?
Non-negotiables include:
Formal verification
Lack of committee
Immutability
Open source code
Solana also benefits from a unique architecture, which can lead to different privacy priorities than its EVM counterparts. Mert points out how, for instance, at ZK compression: “We can actually get mass scale and composability for a privacy protocol on Solana without the need for a rollup. At least not a persistent one when you can use based rollups.”
In the specific case of Solana, the two most relevant verticals for developing privacy are Neobanks and Private DeFi. Nonetheless, in terms of tools and user experience, Solana is still far from a functional, composable private ecosystem.
We also asked Mert his thoughts on the privacy tech stack. As highlighted in this report, privacy should not be taken singularly as a single piece of tech. Instead, we believe that all primitives will interact in the “final privacy stack”. For Mert, the endgame will come when FHE will be combined with ZK. While TEEs and MPCs are pragmatic for certain use cases, they do not provide enough guarantees in an adversarial system.
Last but not least, we took the chance to ask him about Helius Privacy.
It will be developed as a ZK-based UTXO privacy layer on Solana. Helius Privacy will leverage “Zones“, where individual companies can pick their tradeoffs.
A public zone will also be available to all regular consumers, providing full anonymity in an immutable, formally verified manner.
With this context provided, we focus on Solana’s privacy ecosystem and how it addresses various privacy challenges in this section.
Private Compute
There are currently two major providers in this category: Arcium and Magic Block.
Both of them try to solve a similar problem: private computation.
Arcium does it through an MPC to process any data. Arcium splits the data and distributes it across a cluster of independent nodes. These nodes then work together to compute results without seeing the individual inputs. Moreover, while Arcium operates as its own compute network, it settles on Solana, which handles task ordering, network security, and fee payments.
All these private computations take place in Multi-Party eXecution Environments (MXEs), which are customisable, parallelised execution environments.
While Arcium solves the problem of computation, their products also serve privacy on Solana more broadly. They are building a Confidential SPL (C-SPL) token standard, which can enable confidential tokens, transfers, and trading on Solana. We asked the Arcium team a few questions, including where most of their demand comes from. Not surprisingly, for payments and encrypted data analysis, and an increase in institutional demand from healthcare, which enables the training of models on encrypted datasets. C-SPL also enables seamless private transfers, thereby generating institutional interest.
In terms of numbers, Arcium has seen good adoption since its Alpha mainnet launch in early February 2026, processing 900k+ computations and over 3.5 million transactions, with most of this growth occurring at the beginning of May.
Most demand has come from the initial applications, such as Umbra. However, applications such as ZINC and Crafts are going live in the coming weeks, and we expect to see increased demand.
ZINC is doing encrypted proof-of-work mining, and Crafts is using Arcium for sealed-bid auction fundraises, allowing startups to tokenise part of their equity with fair price discovery.
“Some exciting things being built on Arcium that were never done before include capital formation with sealed-bid auctions, opportunity markets, encrypted resolution in prediction markets, and other financial privacy applications.”
Many of these use cases create new and improved markets where the user doesn’t even think about privacy:
Instead, Magic Block solves the privacy-compute problem by utilising TEEs, whereas Arcium MPC relies on cryptographic guarantees. In terms of how their product works, Intel TDX enclaves create a hardware-verified black box called Private Ephemeral Rollup (PER) in which transactions are aggregated and processed, then committed back to Solana.
MagicBlock helps design applications that preserve these properties across the stack, including Confidentiality (protected state), Scalability (high throughput), Composability (still operable with different Solana programs), and Compliance (access-control layer).
While both products solve the problem differently, they help produce private order books, dark pools, and private DeFi rails that can be deployed with minimal code changes. This is evident through the ecosystem building on top of Arcium, for example, covering DeFi, Prediction Markets, Neobanks, and more.
Private Transfers and Balances
With infrastructure for private computations like Arcium and MagicBlock being built out, certain use cases for that infrastructure are beginning to grow as well, including private transfers.
Umbra is the first in the list, built on Arcium’s MPC Infrastructure. Umbra introduces Encrypted Token Accounts (ETAs), a direct counterpart to Solana’s standard Associated Token Accounts, but with balances stored in encrypted form, providing:
Amount Privacy: Transaction amounts are encrypted using the Rescue cypher.
Balance Privacy: Balances are stored as ciphertexts
Linkage Privacy: Linkage privacy is enabled via a shielded pool, with ZK proofs that sever the sender-receiver link entirely.
Apart from this, Umbra offers additional compliance features that enable users to grant auditors and compliance systems selective viewing access without exposing the full transaction history. This feature is important for institutional workflows and for users who want to demonstrate they hold funds without revealing detailed transaction history.
Continuing on the Privacy front, Privacy Cash and Hush are the other two wallet providers.
Privacy Cash uses a Tornado-style shielded pool for SOL: Users deposit SOL to generate a commitment that is added to a Merkle tree, and withdraw to any recipient using ZK proofs that break any onchain link between the deposit and withdrawal addresses.
Hush, on the other hand, is Zcash-inspired but with the addition of DeFi utility. Users can deposit SOL into Hush’s shielded pool, where it gets converted to jitoSOL, passively earning staking yield and MEV revenue while remaining private. Within the pool, users can transfer funds to other Hush participants, receive funds, and transact multiple times without ever touching the Solana public ledger. When they choose to exit the pool, the withdrawal gets unlinked from the original deposit through mixing. The transfer cost within the pool is 0.01 jitoSOL, and a 50 bps fee applies to withdrawals. Hush also integrates Jupiter for private swaps and incorporates geo-blocking for sanctioned regions, giving it a compliance profile among institutional users.
No Onchain Trail
Private transfers fix the problem for companies and institutions which pay their employees or make payments on any private deals onchain but to take it to a deeper level, we need privacy embedded in our daily onchain actions, which is trading. Every order placed on a public AMM is a signal that front-runners, copy traders, and MEV bots can read and exploit, but there are some protocols on Solana fixing it.
First in the list is encrypt.trade, a privacy-first DeFi interface that routes transactions through Jupiter while keeping transaction details private. Additionally, the team adds that “execution quality remains the same as we don’t add any custom routes, we use the same route and liquidity provided by Jupiter”
It wraps the token users wish to swap and encrypts the swap details using ElGamal encryption. The onchain record shows a state change for the wrapped asset type, enough for Jupiter to know it’s routing the correct token. The transaction count, the transactors, the parties, and even whether an execution occurred are processed inside a TEE environment (AWS Nitro Enclaves) and never broadcast publicly. This unlocks private swaps at scale.
Vanish approaches the problem by focusing on a different layer. They are building a private trading infrastructure using shielded transaction routing to protect trading strategies from creating an onchain trail. Rather than wrapping tokens like encrypt.trade, Vanish routes trades through shielded liquidity. Additionally, Vanish has introduced the Vanish Integrity Framework (VIF), which embeds safeguards to prevent the routing of any illicit transactions, powered by Elliptic and Range.
Darklake is another contender in this category, building a ZK-native liquidity infrastructure and dark pool. Its zk-AMM, called “blind slippage pool”, is a commit layer added to the AMM to hide slippage data before execution. Searchers cannot read the order intent before the transaction lands in the block, but they can verify the outcome after the fact. This latency asymmetry prevents sandwich attacks while preserving verifiability. They have extended this model into private perpetual contracts (zk-Perps) with Arcium’s compute layer and Zyga, a reasoning framework that abstracts proof complexity and provides builders with a foundation for secure logic and coordination. Moreover, they recently expanded to become an infrastructure protocol that allows applications and users to verify, connect, and compute privately using their Proof as Intelligence model.
Private Prediction Markets
Private prediction markets are more advanced privacy-enabled applications because users’ strategies are easily copied, causing them to lose their edge. To account for this, protocols are building dark pools specifically for prediction markets using Arcium infrastructure.
Melee is building prediction markets with private order flow enabled. They are doing this by encrypting the order book using Arcium’s MPC infrastructure. Participants can place positions without revealing their direction in the open market until settlement.
The Private AI
As AI agents start to operate more onchain, do every query and every piece of PII they consume become permanently public?
Loyal answers this question through its decentralised, censorship-resistant intelligence protocol, built using Magic Block’s ephemeral rollup execution and Arcium’s encrypted compute. They are building towards onchain AI that protects user data: conversations, queries, preferences, and activity, all stored encrypted on Solana with strict access rules. Users own and can export their encrypted conversation history. They can also self-host their frontend without losing any of their data. Additionally, Loyal supports private transactions and money management, allowing depositors to earn yield privately.
Find the complete report here:
https://docsend.com/v/sjv2g/thefullstackprivacyecosystem
written by @francescoweb3 and @noveleader ✍️
Every week for the last 3 years, we have shared our research for free, directly in your email. Not a subscriber yet? Let’s fix it:
If you are more of a Telegram guy, you can read all of our research without the noise on our TG channel: