Cardano opened this week by finishing last week’s homework. The long-awaited van Rossem hard fork enacted at 21:44:51 UTC on July 18, at the epoch 644 boundary, lifting the protocol’s floor from 10 to 11. That is the same floor we closed the previous issue (#15) on, with the caveat that enactment was still scheduled. What remained once it rose was a quieter question: on top of that floor, who gets the money? Five treasury withdrawals now face a July 28 deadline, and the Constitutional Committee’s membership was settled. Then, midweek, something broke — not Cardano, and not Midnight, but an external bridge. This week moved on three things: the floor, the allocation, and the boundary.
Executive Summary
- The floor rose: van Rossem enacted at 21:44:51 UTC on July 18 (the epoch 644 boundary), taking the protocol to version 11. There was a roughly ten-minute gap in block production immediately after enactment, but it proved transient and normal production resumed in epoch 645.
- The change is a repricing, not a discount: five CIPs introduced new builtins and types, while the cost model for some existing primitives was increased. This was not an across-the-board price cut but a revision of the price list itself.
- Allocation is next: five treasury withdrawals expire at epoch 646 (July 28). Against the 67% DRep threshold required to pass, the leading proposal — Daedalus maintenance — sits at 54.72% as of July 25. None has reached the bar.
- What broke was a bridge: roughly 515 million NIGHT drained from the Cardano side of the third-party Wanchain bridge on July 20. Midnight’s core protocol, validators, and consensus were unaffected. NIGHT ended the week down about 27%.
1. Market Pulse
| Asset | Jul 18 | Jul 25 | WoW |
|---|---|---|---|
| BTC | $63,905 | $64,093 | +0.3% |
| ETH | $1,840.77 | $1,860.44 | +1.1% |
| ADA | $0.16669 | $0.16382 | −1.7% |
| NIGHT | $0.02764 | $0.02018 | −27.0% |
BTC and ETH edged higher, ADA slipped slightly, and the majors spent another week without clear direction. Almost all of this week’s movement sits in NIGHT. Behind that roughly 27% decline are two overlapping forces: the bridge incident discussed below, and distributed tokens continuing to enter circulation. In a week when the index does not move, the one thing that did move deserves careful reading.
2. The Floor Rose — What van Rossem Actually Contains
What changed
van Rossem advances Cardano to protocol version 11 (PV11). It enacted at the epoch 644 boundary, 21:44:51 UTC on July 18.
The contents are less a headline feature than an update to the parts and the price list that smart contracts work with. Five CIPs introduced new builtins and types: modular exponentiation (CIP-109), the dropList builtin for efficient list handling (CIP-132), multi-scalar multiplication over BLS12-381 (CIP-133), an array type (CIP-138), and the MaryEraValue type for multi-asset values (CIP-153). Alongside this, newer builtins became available from Plutus V1, V2, and V3 alike, reducing the need to recompile older contracts purely to reach later primitives.
The direction of the cost model deserves attention. In Intersect’s description, the update includes increases to the cost model of some existing primitives.
increases in the cost model of some existing primitives ——Intersect MBO
So the summary “execution costs went down” is not accurate. More precisely: the available parts expanded, and the price list was redrawn to match. Native array and multi-asset value types genuinely make cheaper what previously required expensive workarounds. But existing patterns were not necessarily left untouched. For developers, this is best read as an update where some things got cheaper and some got dearer, at the same time.
What matters from an SPO’s seat
A hard fork is the work of every node stepping onto the same floor at the same moment. This time there was a gap of roughly ten minutes with no blocks produced right after enactment. It was transient, and epoch 645 returned to normal block production. There was no split and no extended halt.
Tens of thousands of nodes switching together at one predetermined slot, pausing for ten minutes, then running on as if nothing happened — that unremarkable outcome is what a successful hard fork looks like. And this time the decision travelled through the Constitution and the on-chain governance process. Intersect frames van Rossem as a moment in which the decisions around it have increasingly been made through the governance structures established by the Constitution and the on-chain governance framework.
In the previous issue we wrote that the floor was about to move. It moved. From here, the conversation is about what gets stacked on top of it.
3. Allocation Is Next — Five Proposals Expiring July 28
Where the treasury withdrawals stand
What is underway on Cardano immediately after the floor rose is a set of votes about money. Five treasury withdrawals expire at epoch 646 (July 28). Passing requires 67% DRep approval.
| Proposal | DRep yes |
|---|---|
| Se7en Labs: Daedalus Wallet Maintenance and Improvements 2026-2027 | 54.72% |
| Blockfrost’s transformation to not-for-profit | 27.48% |
| Cardano Enterprise Adoption: ticketing platform (4,969,231 ada) | 18.81% |
| Cardano Builder DAO | 7.93% |
| Alchemy by Sundial × Charms: Cardano-native BTC treasury | 6.37% |
(On-chain tally as of July 25, 2026)
Laid out this way, the picture is clear. None of the five has reached 67%. Even the best-supported, Daedalus maintenance, stands at 54.72%. Unless votes move substantially before the deadline, most will lapse rather than pass.
We read this as the threshold working as designed, not as dysfunction. The treasury is not a budget that disburses on application; it is a mechanism through which only proposals clearing a high 67% bar get through. A week in which most proposals fall short is also evidence that the bar is real.
Proposals with later deadlines are moving too. At epoch 647: Bifrost for Bitcoin DeFi (20.99%), the Net Change Limit governing treasury growth (29.27%), Global Order Book (14.57%), and Scalus 2026 (17.23%). At epoch 649 sits AlphaGrowth’s Cardano PRIME (120,000,000 ada, 29.68%).
One clarification worth making: treasury withdrawals are decided by DReps and the Constitutional Committee. SPOs are not voters on this class of action. Where tallies show 0% SPO approval, it is because SPOs hold no vote here, not because they voted against. SPOs do vote on things like hard fork initiation and info actions — and indeed, the info action naming the protocol version 12 hard fork “von Bergen” has 91% SPO approval. The next floor already has a name taking shape.
The Constitutional Committee is settled
DRep voting closed at 21:45 UTC on July 23, producing preliminary results for the 2026 Constitutional Committee election. The four selected are Philip DiSarro (2.0B ada), Leandros BSP (2.0B ada), Marek Mahut / deliberative.cc (1.4B ada), and Cardano Curia (1.3B ada), with the figures indicating the delegated stake behind each. An audit is expected to complete by Monday, July 27, with on-chain submission during epoch 646 (from July 28).
The Constitutional Committee’s role is to assess whether actions conform to the Constitution. It does not decide budgets, but it is the final checkpoint on whether what passes stays inside constitutional bounds. In the same week that allocation went to a vote, the membership of the body that reviews it was fixed. This week’s governance ran on those two tracks side by side.
4. Midnight Watch — What Broke Was a Bridge
What happened
Across nine minutes, from 14:46 to 14:55 UTC on July 20, roughly 515.2 million NIGHT left the Cardano-to-BNB Chain route of the third-party Wanchain bridge in four transfers. That represents about 97.8% of the Cardano-side reserve.
The cause lay in signature verification design. The bridge’s TreasuryCheck validator built its signed message by concatenating fourteen variable-length fields with no separators and no length markers, so field boundaries were not uniquely determined. As a result, a legitimate signature authorizing about 3,110 NIGHT also validated a withdrawal of 203 million tokens — a difference of roughly 65,000 times. Wanchain acknowledged the unauthorized withdrawals and took the bridge offline pending investigation.
The first thing to establish is that this was neither a Cardano nor a Midnight failure; it was an external bridge failure. The Midnight Foundation stated that the incident was isolated to Wanchain’s Cardano-BNB Chain bridge, and that Midnight’s protocol, validator network, and consensus mechanism continued to operate normally and securely.
How the price moved
The drained tokens were sold quickly. Tracing indicates roughly 300 million NIGHT were sold on Cardano decentralized exchanges, of which about 217.7 million were swapped for ADA and about 87.88 million for USDCx.
The price followed that sequence. NIGHT fell from $0.0252 at 16:00 UTC on July 20 to $0.0186 by 20:00 the same day, reaching a weekly low near $0.0174 in the early hours of July 21. It then recovered, reaching $0.0237 by 00:00 on July 22, and closed the week at $0.0202.
One note: a fair number of reports date the incident to July 21. That is the publication date; the transactions themselves occurred on July 20. The price record is consistent with the July 20 drain and the selling that followed it.
How we read it
For a network built around privacy, the incident lands on the point with some irony. What held was Midnight’s core. What broke was an external bridge built to carry assets to it.
The security of a chain and the security of the routes in and out of it are separate problems. From a user’s perspective the outcome — assets gone — may look the same, but a different cause calls for a different response. Which bridge you cross has become a practical decision carrying the same weight as which chain you use. That clarity of demarcation is what this incident leaves behind.
Midnight itself published its July State of the Network the same week. Alongside policy engagement and educational content, it notes partnerships with Token Terminal and Dune. Being able to follow network activity from external analytics platforms makes it easier to confirm, from the data side, that a privacy-oriented design is not about being invisible but about choosing what to reveal.
5. Risk Dimensions
| Area | This week | What to watch |
|---|---|---|
| Protocol | PV11 enacted; normal after a ~10 minute gap | Whether contracts using the new builtins actually appear |
| Execution cost | Some primitives repriced upward; new types create headroom | Which way real costs move for major DeFi |
| Governance | Five treasury items expire July 28; all below 67% | Vote movement before the deadline, and resubmission after failure |
| Oversight | Four CC members set; audit due July 27 | Completion of on-chain submission (epoch 646) |
| Boundary (bridges) | Large drain on a third-party bridge; bridge offline | Recovery conditions and audits of similar signature schemes |
| Distribution (Midnight) | NIGHT −27% on the week, partial recovery post-incident | Pace of circulating supply growth and real demand |
6. What to Watch Next Week
- Resolution of the five treasury items — how DRep votes move into the July 28 deadline, and whether anything passes or lapses.
- Constitutional Committee finalization — audit completion by July 27 and on-chain submission in epoch 646.
- PV11 in practice — whether implementations actually using the new array type and BLS12-381 builtins emerge.
- Bridge aftermath — Wanchain’s findings and recovery conditions, plus audits of other bridges using similar signature verification.
- NIGHT supply and demand — whether the post-incident recovery holds against continuing circulating supply growth.
Conclusion
This week Cardano raised its floor, moved into votes about allocation on top of it, and watched an external bridge fail. van Rossem enacted on schedule and, after a ten-minute gap, ran on without incident. But “execution costs went down” is not an accurate summary — the price list was redrawn. And the five treasury proposals have all yet to reach 67%.
In the previous issue we wrote that the next question was what gets stacked on top. This week that question came back in a concrete form: a vote. What gets stacked is decided not by the floor, but by the ballot.
Sources
- Intersect MBO — Weekly Update #121 (2026-07-24; enactment confirmation, CC election results, proposal deadlines): https://www.intersectmbo.org/news/intersect-weekly-update-121-july-24-2026
- Intersect MBO — Cardano upgrade: van Rossem Hard Fork: https://intersectmbo.org/news/cardano-upgrade-van-rossem-hard-fork
- Hard Fork to Protocol Version 11 “van Rossem” (governance action): https://gov.tools/governance_actions/fdd468da5cc4ac8431dcd7e2b3211666c73bc229f85879469f67f1d9d51d344d
- Daedalus 11.1.0 release notes (2026-07-22; Mithril sync for existing installs): https://github.com/input-output-hk/daedalus/releases/tag/11.1.0
- Technical analysis of the Wanchain bridge exploit (signed-message encoding flaw): https://www.cryptotimes.io/insights/wanchain-night-bridge-exploit-signature-flaw/
- NIGHT rebound and the wider industry question (CoinDesk, 2026-07-22): https://www.coindesk.com/business/2026/07/22/midnight-token-rebounds-after-wanchain-bridge-hack-hoskinson-calls-for-industry-overhaul
- Midnight — State of the Network, July 2026 (Token Terminal / Dune): https://midnight.network/blog
- On-chain governance tallies: Koios (https://api.koios.rest)
- Prices: CoinGecko (https://www.coingecko.com/)
Transparency Note
Prices are CoinGecko daily snapshots (UTC basis), comparing July 18 and July 25. The previous issue used a snapshot taken at a different moment, so its July 18 figures differ slightly. Intraday NIGHT movement is likewise from CoinGecko hourly data. Governance approval rates are Koios tallies as of July 25, 2026; voting continues until the deadline, so these are not final figures. The Constitutional Committee selection is preliminary, with audit and on-chain submission still outstanding. Reported losses from the bridge incident range from roughly $9 million to $13 million depending on the source, so this article uses token quantities instead.
This article is provided for informational purposes and does not constitute a recommendation to buy, sell, or hold any crypto asset.