Privacy ecosystem map v1: private money to confidential compute
Thread2,699 words97 likes
Privacy asset map with mcap/FDV, 7d and 30d moves and ATH distance. Notes STRK up 99% in 7d on the quantum trade and Arcium's ARX trading 370x below ZEC and 19x below VVV on mcap.
Key claims
ZEC $20.8B with 4.96M ZEC shielded (29.2% of supply)
STRK +99% 7d, +254% 30d
ARX 370x below ZEC and 19x below VVV on mcap
Privacy is still one of the hottest narratives out there, and there is likely even more growth ahead given many project have stacked roadmaps.
There is still a lot of opportunity if you pick the right plays.
So here's my ecosystem map (mcap/FDV, 7d and 30d moves and distance from ATH):
1) Private money
- $ZEC ($20.8B, -6% 7d, +7% 30d, 62% off ATH):
Zcash is optional privacy: you move coins into a zk shielded pool and from there nobody sees sender, receiver or amount. For years the problem was that hardly anyone bothered to shield. That changed: 4.96M ZEC sit shielded now, 29.2% of supply and an ATH, 4.1M of it in Ironwood, the new pool from July that Brave made its default last week. On top, NU7 is on testnet since 4 Oct, the team presented Tachyon this week (recursive SNARKs, ~150 TPS target), and ZCSH on NYSE Arca is the first spot ETF for a privacy asset anywhere. On quantum, hash-based signatures for the transparent pool are targeted for January, while the shielded pool needs its own migration (Epoch, end of 2027). There are no programs inside the pool tho. ZEC is private money and nothing else, which is also why it stays the asset the whole vertical is priced off.
→ DYOR: @Zcash
- $XMR ($10.0B, -5% 7d, +3% 30d, 34% off ATH):
Privacy by default instead of by choice. Ring signatures, stealth addresses and RingCT hide sender, receiver and amount on every single tx, no opt-in. The weak spot was always the ring itself: 16 members, so you hide among 15 decoys and a good heuristic narrows that down. FCMP++ replaces the ring with a proof that you own one of the 150M+ outputs on the entire chain, which makes the whole history your anonymity set. The third beta stressnet has been live since 5 Oct, hardware wallets and multisig don't work on it yet, and there is no mainnet date. Add that it's delisted on most regulated venues and has no ETF, and you get why it trades like ZEC's discount twin. FCMP++ is the catalyst, the missing listings are probs the ceiling.
→ DYOR: @monero
2) Private transfers and payments
- $RAIL ($161M, -5% 7d, +38% 30d, 53% off ATH):
Railgun has been live since 2021 and is still the most used private DeFi rail on Ethereum. It's a zk shielded pool on Ethereum, Arbitrum, BSC and Polygon, and the trick is that you call DeFi from inside the pool, so Uniswap or Aave never learn who you are. Private Proofs of Innocence let you show your funds are clean without doxxing them, which is what keeps it usable post Tornado. Blockworks just put a dashboard on it: $4.6M annualised revenue, staking yield up from 4.3% to 9.1%, weekly txs at an ATH of 5K.
→ DYOR: @RAILGUN_Project
- $XAN ($28M mcap, $114M FDV, -7% 7d, +1% 30d, 96% off ATH):
Anoma does privacy as a consumer product. The architecture is intent-centric (you say what you want, solvers figure out how, privacy sits in the protocol), but what you actually touch is AnomaPay. V1 is live on Ethereum plus five EVM chains with 20 assets: shielded transfers, private yield accounts on Morpho, confidential swaps via Bebop, private payroll, private tokenized stocks on BNB, and since 6 Oct native private swaps. For normies this is probs the right abstraction, an app rather than a chain. It's also the cheapest optionality in the basket, bc it sat flat while everything around it ran (take that as you will).
→ DYOR: @anoma
- $ZEN ($119M mcap, $136M FDV, -6% 7d, -3% 30d, 96% off ATH):
Horizen went the compliance route. It's an EVM-native L3 on Base, and Vela, its confidential compute layer, lets an app keep its state private while disclosing exactly what a regulator asks for and nothing more. That is selective disclosure built for the EU and UK transfer rules, not anonymity, and it's a very different buyer than the ZEC crowd. 21M capped supply, ERC-20 on Base, staking live since July, a token-weighted Special Council elected two weeks ago, first Vela accelerator cohort wrapping up this month. It has lagged ZEC the entire run, which you can read either way.
→ DYOR: @horizenglobal
3) Programmable privacy chains
- $AZTEC ($51M mcap, $176M FDV, -2% 7d, +17% 30d, 57% off ATH):
Aztec is the purest ZK answer to what Zcash can't do, which is programs. Private smart contracts written in Noir, proofs generated on your own device so the network never sees your state, decentralised sequencers and provers underneath. It had its hiccups, no point pretending otherwise: Alpha V5 shipped a critical proving bug (found in July, disclosed in Aug). Since then the core of the proving system got formally verified (1 Oct) and the V6 code is ready and in governance to head to mainnet since 9 Oct. Galactica is building the first private AI inference app on it. On quantum, Zac called bunker mode hyperbolic and warned against rushing into untested crypto, so the security roadmap is still open. It's the only L2 doing full stack programmable privacy, which makes it both the highest beta and the highest execution risk in this section.
→ DYOR: @aztecnetwork
- $STRK ($760M mcap, $1.0B FDV, +99% 7d, +254% 30d, 98% off ATH):
Two weeks ago this was a dead L2 token that happened to have a use case. The use case is STRK20, privacy at protocol level: shield and unshield in your wallet, private transfers, private swaps on Avnu against the same liquidity as public markets, private payroll on DashX, private staking of strkBTC via Endur, even private Polymarket trading. Selective disclosure with viewing keys rather than anonymity, a.k.a. the version institutions can actually use. Then the quantum trade landed on top. Starknet said it is "actively considering becoming an L1" to be fully post-quantum by 2027, STARKs are hash-based already, and a phone signed a testnet tx with an ML-DSA key without a hard fork. That's a consideration tho, no proposal, no vote, and a 2x week already prices a lot of it. Now it's the privacy plus quantum beta, size it like one.
→ DYOR: @Starknet
- $ROSE ($70M mcap, $87M FDV, +1% 7d, +29% 30d, 99% off ATH):
Oasis has been running a confidential EVM chain since before it was a narrative. Sapphire keeps contract state and inputs encrypted even from node operators via SGX with remote attestation, and ROFL runs off-chain apps in SGX and TDX enclaves that Sapphire manages (bots, oracles, agents). You can deploy Solidity to it today. The team is the quietest in the basket (one substantive post in a month) and the token is 99% off ATH, so it is either the cheapest programmable privacy chain on this list or a dead project with working tech. Probs a bit of both.
→ DYOR: @OasisProtocol
- $CC ($4.7B, -3% 7d, +20% 30d, 39% off ATH):
Canton is the chain Wall Street actually settles on, and privacy is the reason. Need-to-know visibility, sub-transaction privacy, a synchronizer that orders transactions without ever seeing the terms. September alone brought Franklin Templeton and TreasurySpring as super validators, 10 new featured apps incl. Circle, and 442.6M CC burned through traffic. Since then LSEG became a super validator too (its DiSH already settles intraday repos on Canton), SoFi added native CC, and a Universal Burn CIP is in draft to route all synchronizer fees through CC burns. DTCC's tokenization service is the next shoe. Not privacy for CT, privacy that institutions pay for, in CC.
→ DYOR: @CantonNetwork
- $MON ($295M mcap, $2.5B FDV, -23% 7d, +6% 30d, 49% off ATH):
Monad wants the same lane. Private Settlement, announced 6 Oct, is a design for sovereign domains on the L1 where transactions, state and contract code are encrypted, even from the chain itself. The institution picks its validators and jurisdiction, access is denied by default with scoped access for auditors, and every domain still settles atomically against Monad's pooled liquidity. No fork, and no offchain components. Target customers are banks, treasuries, asset managers and OTC desks, a.k.a. the Canton pitch on a general purpose L1. There's also a post-quantum roadmap from 28 Sep with address-preserving key rotation, wallet auth and consensus in progress, encrypted mempool last.
→ DYOR: @monad
1/3
4) Confidential compute (MPC, FHE, TEE)
- $ARX ($56M mcap, $268M FDV, -4% 7d, +88% 30d, 42% off ATH):
My main bet, so I'll keep it short. Arcium is an MPC network where nodes compute on data none of them can see, settled on Solana. Encrypted compute is live with five apps on mainnet (Umbra, Zinc, Bench, Rips, Crafts). C-SPL is coming: encrypted balances for any SPL token and programs that can own confidential accounts, a.k.a. the Zcash use case with programmability and composability on top. Blackthorn got its first real update on 8 Oct: the CPU TEE is out of the stack, the GPU's session keys are held as MPC shares, the host is untrusted and the Nvidia GPU runs the model in CC mode, so nobody sees plaintext even with physical access. That's basically the Venice use case with a better trust model. Neither product is fully live yet tho. However, ARX still trades 370x below ZEC and 19x below VVV on mcap. That gap is the trade.
→ DYOR: @Arcium
- $ZAMA ($194M mcap, $866M FDV, -13% 7d, +63% 30d, 29% off ATH):
FHE means computing on encrypted data without ever decrypting it, and Zama is the one that turned it into product. ERC-7984 confidential tokens are live on Ethereum mainnet: cUSDC, cUSDT, confidential Morpho vaults, confidential incentives on Merkl. GSR did the first confidential OTC trade on it, Bron ran confidential payroll, and it is the default confidentiality layer for T-REX Ledger. Shielded TVL passed $100M on 5 Oct, up from $85M two weeks earlier. The tradeoff is real tho: FHE is slow, and shared state still needs a threshold committee to decrypt. Still, it is the one confidential DeFi stack institutions are actually using today.
→ DYOR: @zama
- $NIL ($44M mcap, $88M FDV, +5% 7d, +105% 30d, 90% off ATH):
Nillion is making the most interesting pivot in the basket. The base layer is blind computation, data split across nodes and computed on with MPC and HE so no node ever holds plaintext. The product on top is Encrypted Markets, built around Covenants: an encrypted instruction that decrypts itself when its condition is met, with no operator holding the key. Dusk, the first phase, went live on Ethereum mainnet on 6 Oct, Blacklight L1 nodes are up on Ethereum (70k NIL min stake, rewards from 0.5% inflation plus trigger fees), Darkfall extends Covenants across the EVM next, Hyperliquid integration planned.
→ DYOR: @nillion
- $PHA ($51M mcap, $60M FDV, -13% 7d, +128% 30d, 96% off ATH):
Phala is the hardware side of private AI. Phala Cloud runs private LLM inference, sealed agents and GPU jobs on Intel TDX plus Nvidia H100, H200 and B300 confidential GPUs, attestation included, 5k+ users, SOC 2. It is one of the two enclave providers behind Venice's end to end encrypted mode, and the execution layer under Rift, the Paradigm-backed onchain execution API that launched 1 Oct. Migrated from Polkadot to Ethereum last Nov. The tradeoff is the same as for every TEE: you trust the chip/vendor.
→ DYOR: @PhalaNetwork
5) Private AI
- $VVV ($1.1B mcap, $1.8B FDV, -20% 7d, -10% 30d, 35% off ATH):
Venice is the consumer front end of private AI. In end to end encrypted mode your prompt is encrypted on your device, decrypted only inside a GPU enclave at NEAR AI Cloud or Phala, and you get an attestation proof back. Claude Opus, Sonnet and Haiku 5.5 are all available anonymously now. Staking gives you a pro-rata share of API capacity and mints DIEM, $1 of API credit per day in perpetuity. Emissions stepped down from 2.5M to 2M VVV a year on 1 Oct, the largest discretionary burn to date ($414K) ran on 7 Oct, it got listed on Solana via Sunrise, and the $65M Series A at $1B closed in July. The only name here with revenue at scale. Trust model: the chip and the vendor. -20% on the week with no news it could be a dip worth watching if you believe in the narrative.
→ DYOR: @AskVenice
- $NEAR ($7.0B, +14% 7d, +114% 30d, 74% off ATH):
NEAR is the rail under Venice and the chain that actually did the quantum homework early. NEAR AI Cloud runs inference in hardware enclaves with Intel Trust Authority attestation, is a provider on OpenRouter now, and Venice chose it for encrypted inference. On near(.)com deposits are confidential by default, you hold stocks, gold and perps confidentially, and Hyperliquid perps run with the link between funding wallet and trading account hidden. On quantum: anyone can add an ML-DSA key to their account on mainnet in one tx and keep the account, NEP-655 hash-committed accounts are on testnet with mainnet in weeks, and a Q-day recovery registry for other chains is in the works. Plus a live proposal to cut issuance from 2.5% to 1.6%. Not cheap after +114% in a month, but it's the only large cap where privacy and post-quantum are shipped rather than promised.
→ DYOR: @NEARProtocol
- $RLC ($86M, +169% 7d, +247% 30d, 94% off ATH):
iExec has been a confidential computing marketplace since 2017 and now sells itself as the rail for confidential AI. Apps and agents run inside TEEs on Arbitrum, DataProtector decides who may process a dataset without ever revealing it, Web3Mail lets you message a wallet without learning its email. The framing is correct, encryption at rest and in transit is solved and in use is not, but there is no new product milestone this month. The tripling came on the narrative and an Aster perp listing.
→ DYOR: @iEx_ec
6) Private trading venues
- $DIME ($47M mcap, $70M FDV, +82% 7d, +670% 30d, 5% off ATH):
Paradex runs privacy perps on its own Starknet-stack L2. Orders, positions, PnL and liquidation levels are hidden, the RPC only hands your data to you after a signed request, and state goes to Ethereum as encrypted diffs with ZK verification. The caveat: a three member Privacy Council (Paradex Foundation, Paradex, Karnot) holds the decryption keys as an escape hatch. Zero fees, unified margin, and options are the story now: 20% of onchain options notional since the Paradigm RFQ launch, $430M options volume in three weeks, OI past $550M, four new RWA markets. Nobody can read your book, so nobody can trade against it, and no other live perp DEX with a token offers that. Up 8x in a month so the easy part of PA is done, but it still sits at a fraction of the FDV of any perp DEX running similar OI.
→ DYOR: @paradex
- $EDEL ($27M mcap, $39M FDV, -15% 7d, +245% 30d, 65% off ATH):
Edel is the Canton-native take on the same idea. Edel Markets is the perp DEX, not live yet, with positions and orders private by default through the network's need-to-know model instead of bolted on. What is live is Edel Stocks: SPYe, QQQe, NVDAe and SPCXe, issued by Edel itself as an approved Canton asset issuer, trading on Cantex when Wall Street is closed, $3M+ cleared on the two ETFs in under ten days. Worth knowing what you hold tho: these are 1:1 economically backed debt instruments, not share claims, and only KYB'd institutions can mint or redeem them in USDCx. EDEL staking routes part of the revenue to holders in CC, tiers announced for this month. Seat in the DTCC Digital Assets Industry Working Group, and a Swiss team. It's the only tokenised Paradex peer, bc Ekiden and Hibachi, the other private-by-default venues, have no token yet.
→ DYOR: @edeldotfinance
2/3
P.S.: Currently (still) betting on NEAR, ARX and AZTEC, did hold ZAMA and EDEL but rotated (still watching both closely), missed the recent STRK and DIME pumps but keeping them on the radar, and am generally trying to keep an eye on the broader vertical.
All NFA tho ofc. DYOR.
3/3
