Why Midnight needs Cardano: the token, the validators, and the technical debt

Midnight is a blockchain of its own. It has its own ledger, its own nodes, its own smart-contract language and its own consensus. Yet almost everything that makes it run today starts on another chain. Its token was born on Cardano, most of that token still lives there, the fees on Midnight are paid with a resource generated from Cardano balances, and the validators Midnight plans to rely on are Cardano's stake pool operators.

That is not an accident of history. Midnight is built as a partner chain to Cardano: a separate blockchain that borrows Cardano's token infrastructure, its stake and, in time, its block producers, instead of building all of that from scratch. This post explains what Midnight is, where exactly it leans on Cardano, and which parts are still unfinished. We call those parts its technical debt.

What Midnight is, in one minute

Midnight is a privacy-focused blockchain. Most chains are fully public: anyone can read every balance and every transaction. On Midnight, a contract's state is split in two. The public half lives on the chain. The private half stays on the user's device. A zero-knowledge proof (a short cryptographic proof that a computation was done correctly, without showing its inputs) ties the two together, so validators can check that the rules were followed without seeing the private data. Developers decide what becomes public, and must mark it explicitly. Our earlier deep dive on how Midnight works covers that model in detail.

Midnight produced its genesis block on 30 March 2026. Since then, blocks have been produced by nine federated partners, among them Google Cloud, Blockdaemon, MoneyGram, Worldpay, eToro and Vodafone's Pairpoint.

Why a separate chain at all

If Cardano is meant to secure Midnight, why not build privacy into Cardano itself? Cardano founder Charles Hoskinson has given a direct answer. "The technology in Midnight is more sophisticated and riskier than in Cardano. If we tried to bring that down to the basic level of Cardano, it would create a very bloated protocol," he said in June, as reported by KuCoin News.

The partner-chain design is the compromise. Cardano's base layer stays simple. The heavy machinery, zero-knowledge circuits, a private state model, a new contract language, lives on Midnight, where a failure would not take Cardano down with it. In return, Midnight does not start from zero: it can use what Cardano already has. Midnight's own tokenomics post from June 2025 says the network "will leverage Cardano's robust proof-of-stake system for consensus and security".

Where Midnight leans on Cardano

1. The token was born on Cardano

NIGHT, Midnight's token, launched on 4 December 2025, almost four months before Midnight's own mainnet existed. It launched as a Cardano native asset: a token that lives directly on Cardano's ledger, without a smart contract wrapping it. The fixed supply is 24 billion NIGHT.

When Midnight's mainnet arrived, that supply was to be mirrored onto Midnight's ledger, with a rule that any given token can be unlocked on only one of the two chains at a time. That rule matters: it stops the same NIGHT from being spent on both chains at once.

Today the two forms are usually called cNIGHT (NIGHT on Cardano) and mNIGHT (NIGHT on Midnight). Exchanges, according to community guides, pay out withdrawals as cNIGHT to Cardano addresses only.

2. Fees on Midnight can be paid from Cardano

Midnight does not charge fees in NIGHT. Holding NIGHT generates DUST, a shielded resource that pays for transactions and refills over time. DUST cannot be sent or traded.

The interesting part is that NIGHT does not have to move to Midnight to do this. Midnight's own open-source app lets a holder register a link between a Cardano address that holds NIGHT and a Midnight address. Eight Plutus contracts (Cardano's smart-contract system) store those registrations on Cardano. Midnight reads them and credits DUST to the linked Midnight address. In other words: a balance that sits on Cardano powers transactions on Midnight.

3. The bridge runs on Cardano's clock

For NIGHT that should actually live on Midnight, there is a protocol-level bridge, specified in Midnight Improvement Proposal 20 (MIP-20). The design locks cNIGHT in contracts on Cardano and mints the same amount of mNIGHT on Midnight, so the total never exceeds 24 billion.

The bridge waits for Cardano. Before it mints, it lets about 12 hours pass so the Cardano transaction is effectively irreversible. The proposal names the reason, the risk that Cardano reorganises recent blocks, and calls the result "a significant transaction finality time". Community guides describe the bridge as one-way for now: NIGHT can move from Cardano to Midnight, not back.

4. The validators are meant to be Cardano's

Midnight's documentation describes the starting point plainly. The first validator set is permissioned nodes run by Federated Node Operators, and Cardano stake pool operators (SPOs) are "supported at a later date". A single setting, the D parameter, controls the split between those permissioned validators and registered ones, so decentralisation is a dial that can be turned gradually rather than a switch.

Even today, every Midnight node watches Cardano. The docs list a persistent connection to Cardano, through a database filled by cardano-db-sync (a tool that copies Cardano's chain into PostgreSQL), as a requirement for running one. That is how a partner chain learns which stake pools have registered and what their stake is.

Hoskinson has said that SPOs and the people who delegate to them will earn both ADA and NIGHT when they take part, and that Cardano's treasury is set to collect fees from partner chains in their own tokens, according to a January report by The Crypto Basic. None of that applies yet: no SPO is producing Midnight blocks.

What Cardano gets back

The dependency runs both ways. Cardano gets a privacy layer it chose not to build into its own base layer, a second income stream for its stake pools, and, if the treasury plan goes ahead, fees from activity on another chain. Midnight gets a token distribution, an existing set of professional validators and a security model it does not have to bootstrap.

The technical debt

Every young network ships with parts that are good enough for now and will have to be replaced. That is what we mean by technical debt here: the term is ours, not Midnight's, and each item below is a documented fact rather than a criticism.

Federated block production

Six months after genesis, the nine launch partners still produce every block. Moving to Cardano SPOs is reported as the next phase of Midnight's roadmap, but no date has been confirmed by Midnight. Until it happens, the "Cardano security" in the design is a plan, not the present. The security Midnight has today is the security of nine named companies.

A bridge with one door and a long wait

The protocol bridge works in one direction and takes about 12 hours. That is a deliberate safety choice, but it shapes how people use NIGHT. Most liquidity stays on Cardano, where exchanges pay out. Anyone who wants NIGHT on Midnight waits half a day, and cannot yet bring it back.

That gap has already been filled by others. In July, an exploit of Wanchain's third-party bridge between Cardano and BNB Chain drained hundreds of millions of NIGHT. CoinDesk reported 290 million; other tallies are higher. The NIGHT price fell about 43% before partly recovering, per CoinDesk. The Midnight network itself was not affected, but holders were: a bridge outside Midnight's control carried the token Midnight depends on.

Two forms of one token

cNIGHT and mNIGHT are the same asset with different behaviour, held in different wallets, and tied together by registrations and a bridge that both live on Cardano. It works, but it is a lot of moving parts for one token, and each of them is a place where users can get confused or where software can fail.

A heavy node

Running a Midnight node means running Cardano infrastructure too: the docs require the cardano-db-sync database next to the Midnight node. For the stake pool operators Midnight wants to attract this is familiar ground. For anyone else it raises the cost of running a node.

Tooling that is still settling

Zero-knowledge apps need a proof server: a program that computes the proofs, usually on the user's side. Midnight's proof server has had a rough release cycle. A change merged into Midnight's ledger repository on 1 October says plainly: "A number of recent releases left the proof server release in a broken state, because of issues with manifest signing." A docs change merged on 22 September pinned the recommended proof server to version 8.1.0, because the latest tag lagged behind.

The same docs change notes that calls from one contract to another are supported from toolchain 0.33.0 with ledger 9. That is a newer ledger than the 8.x line the docs point developers to today. Contract-to-contract calls, and private custom tokens, are reported to arrive later this year. Until then, each contract on Midnight largely stands alone.

Fast patching

The September network update lists node version 1.0.3, a security patch, and reminds validators to upgrade their binaries before runtime updates are applied. Frequent, coordinated upgrades are normal for a young chain and easier with nine known operators. They will be harder once block production is open to hundreds of stake pools.

How to read this

Midnight's dependence on Cardano is the design, not a flaw in it. A partner chain is supposed to borrow. The open question is timing. The parts Midnight borrowed are already in place: the token, the fee registrations, the bridge, the Cardano data feed into every node. The part that would make the borrowing pay off, Cardano's stake pools actually securing Midnight, is still to come. Until it does, Midnight has Cardano's dependencies without yet having Cardano's decentralisation.