On June 23, 2026, the Ethereum Foundation announced what it called a “sweeping reset”: 54 employees terminated — roughly 20 percent of total staff — the Privacy and Scaling Explorations lab shut down, and a 40 percent budget reduction taking effect immediately. Within the same week, co-executive director Hsiao-Wei Wang stepped down, following the departure of her co-director Tomasz Stańczak in February. Nine senior figures have left the organization since January. The restructuring that followed reorganizes the remaining team into five domain clusters under an interim leader, with no announced timeline for a permanent replacement.
The market read this as a maturity signal. Ethereum co-founder Vitalik Buterin framed it the same way, describing the shift to an endowment-style operating model — targeting a 5 percent annual spend rate by 2030, down from the roughly 15 percent the EF was running — as the difference between a sustainable institution and one burning down its runway. Coverage generally followed that frame: EF getting leaner and more focused, a deliberate evolution toward sustainable nonprofit governance, the kind of organizational discipline that long-term institutions require.
The specific details of what was cut, who left, and what the endowment math actually requires suggest a more complicated picture. The EF’s announcement is internally consistent as a governance document. The gap between what it says and what the specifics imply is the part worth examining — because the implications play out over a timeline that is concrete enough to monitor, and the questions they raise have answers that will become visible within 12 to 18 months.
It is also worth being precise about the scale of what June 23 represents in the EF’s history. The EF has run restructurings before — grant program resets, leadership transitions, research mandate adjustments. None of them combined a 40 percent budget cut, a 20 percent headcount reduction, the closure of the organization’s primary applied cryptography unit, and the departure of both co-executive directors within the same six-month window. The combination is the signal, not any single element in isolation.
What PSE Actually Did — and Why Its Closure Is Not Incidental
The Privacy and Scaling Explorations team — commonly abbreviated PSE — was the Ethereum Foundation’s applied cryptography unit. The distinction between “applied” and “theoretical” matters here. PSE was not producing academic papers about zero-knowledge proofs in the abstract. It was building production-grade cryptographic tooling for real applications on Ethereum.
That tooling included MACI, a protocol for on-chain voting that prevents coercion by making individual votes cryptographically private while keeping aggregate outcomes publicly verifiable. It included Semaphore, a framework for anonymous credentials that allows users to prove membership in a group without revealing which member they are — the underlying privacy layer for applications like whistleblower systems, anonymous polling, and dark pool order matching on-chain. It included PlasmaFold, an approach to privacy-enabled Layer 2 transfers. And it included what PSE called “prove anywhere” research: making zero-knowledge proof generation practical on consumer devices rather than requiring specialized hardware or server-side computation.
The Ethereum Foundation’s restructuring consolidates research under a new mandate called CROPS: censorship resistance, resilience, openness, privacy, and security. The P in CROPS is “privacy.” The organization that was doing the applied-cryptography work for Ethereum privacy — MACI, Semaphore, the consumer device ZK work — was disbanded the same week this mandate was announced. The Protocol Cluster’s documentation describes L1 privacy as a “long-horizon goal.” It does not name who executes it. That gap is not a technicality; it is a resourcing decision presented as a strategic direction.
Zero-knowledge proofs are the technology Ethereum’s scaling roadmap has been built around for three years. EIP-4844 reduced L2 costs by providing blob data availability. The next layer of Ethereum’s scaling plan requires ZK proving systems that can run on mainstream hardware at consumer speeds. That is precisely the category of work PSE was developing. The EF’s restructuring announcement treats PSE’s closure as a budget rationalization. It is also a research capacity decision with a specific roadmap implication that the announcement does not address.
The Ethlabs Formation — and the Bitcoin Development Analogy
Five former Ethereum Foundation researchers — Ansgar Dietrichs, Barnabé Monnot, Caspar Schwarz-Schilling, Josh Rudolf, and Julian Ma — launched Ethlabs in June 2026 as an independently funded, nonprofit research organization focused on Ethereum’s development. Ethlabs has secured backing from Joseph Lubin, Bitmine, and Sharplink. Its research agenda is oriented toward what it calls the “15-minute finality problem” and institutional adoption — how Ethereum’s consensus mechanism can be hardened to the point where institutional market participants can rely on finality guarantees comparable to traditional settlement systems.
Proponents of the EF’s restructuring point to Ethlabs as evidence that talent leaving the EF does not mean talent leaving the Ethereum protocol space. The argument continues: Bitcoin development has been distributed across multiple independent organizations — Chaincode Labs, Spiral (a subsidiary of Block), Brink, and others — for years, without a large centralized foundation. Bitcoin is widely considered the more resilient network precisely because its development is not concentrated in a single organization that can be restructured, underfunded, or mismanaged.
This analogy is instructive but incomplete in a specific way. Bitcoin’s distributed development model functions because Bitcoin’s protocol is intentionally conservative. Bitcoin Core changes slowly by design — the social consensus for protocol changes is deliberately high, and the network has reached a state where the primary ongoing work is maintenance, optimization, and modest additions through soft forks. The development model matches the protocol’s rate of change.
Ethereum’s protocol is the opposite. It changes fast, requires rapid coordination across the base layer and multiple L2 implementations simultaneously, and is in the middle of a multi-year roadmap (The Surge, The Scourge, The Verge, The Purge, The Splurge) that requires synchronized upgrades. The Ethereum Foundation has historically been the coordination mechanism for this — the organization that holds the institutional memory of upgrade decisions, manages the All Core Devs call process, maintains the EIP repository, and provides the continuity that distributed teams need to align on. Ethlabs fills a specific research gap around institutional finality. It is not a coordination mechanism and does not perform the functions the EF has been performing for protocol-wide upgrades.
The Ethereum L2 economics in 2026 show that Arbitrum, Base, and Optimism operate with substantial research and development budgets of their own — they are not dependent on EF-funded work for the features they ship. That segment of the network may be relatively insulated from the EF’s cuts. The L1 protocol development coordination is a different question, and the EF’s restructuring concentrates remaining capacity on narrower goals at a lower funding level than it has operated at in years.
The Endowment Math Assumes a Stable Treasury
The financial argument for the EF’s restructuring is straightforward: spending 15 percent of treasury assets per year is how a nonprofit runs out of money in seven years. Spending 5 percent per year produces a theoretically indefinite runway. Universities and museums operate this way. The EF is now planning to operate this way. This is, in accounting terms, correct.
The EF’s treasury is predominantly held in ETH. The endowment math — how much the EF can spend in year five of the new model — depends entirely on what the ETH treasury is worth in year five. A 5 percent spend rate on a treasury worth $2 billion is $100 million annually. A 5 percent spend rate on a treasury worth $1 billion is $50 million annually. A 5 percent spend rate on a treasury worth $500 million is $25 million annually. These are materially different research budgets, and the difference is determined by ETH price performance over the intervening years, not by any governance decision the EF makes.
The institutional flow data from June 2026 provides relevant context. Spot Ethereum ETFs experienced a sustained underperformance relative to Bitcoin ETFs throughout the month. While Bitcoin ETFs saw $4.33 billion in outflows over a 13-day streak before partially recovering — with BlackRock’s IBIT stabilizing and leading an $86 million inflow day — Ethereum ETF outflows continued structurally, with BlackRock’s ETHA recording negative flows even on days when IBIT turned positive. The divergence between institutional Bitcoin demand and institutional Ethereum demand is not a pricing artifact. It reflects a specific institutional judgment about near-term fundamentals — one that directly affects the EF’s treasury value.
The ETHB institutional yield gap has been a persistent structural feature of how institutional allocators approach Ethereum versus Bitcoin. Institutional ETH products that cannot offer staking yield — because of SEC restrictions on the currently approved ETF structures — are competing on a disadvantaged basis against platforms where ETH yield is accessible. The EF’s endowment model inherits this structural dynamic. If ETH/BTC compression continues through 2027 and 2028, the endowment model will produce a smaller EF in real terms even if the 5 percent governance target holds perfectly.
Leadership Continuity and the Coordination Risk
The departures are specific enough to warrant naming. Tomasz Stańczak, co-executive director, stepped down in February 2026. Hsiao-Wei Wang, the other co-executive director, resigned on June 18 — five days before the restructuring announcement was published. Bastian Aue has assumed expanded responsibilities in an interim capacity. No timeline for a permanent leadership appointment has been announced publicly, and no search process has been described.
The absence of a succession plan is itself a data point. Major nonprofit institutions facing significant restructuring typically announce interim leadership alongside a timeline for permanent placement — both because the timeline anchors expectations internally and because it signals to external stakeholders that the governance transition is managed rather than reactive. The EF’s announcement does neither. Whether this reflects deliberate optionality in the leadership selection process or organizational uncertainty about what the permanent structure should look like is not clear from the available information.
Nine senior figures have left since January 2026. The restructuring into five protocol clusters requires each cluster to have effective leadership with clear mandates and the institutional memory to make decisions across competing priorities. The EF has just redistributed responsibilities to a workforce 20 percent smaller than it was six months ago, led at the top by someone who has not yet been given a permanent appointment. Whether the clusters have the leadership depth to function effectively is a question that will be answered by the next upgrade cycle, not by the restructuring announcement.
The next major Ethereum upgrade after Glamsterdam is expected to address the finality timing problem that Ethlabs is researching independently. The protocol research for that upgrade — the work that previously would have involved EF researchers in the All Core Devs process — will now involve researchers at Ethlabs, researchers at L2 teams, and an EF team that is smaller and in organizational transition. Whether that produces slower coordination, worse-coordinated upgrades, or no meaningful change in output is a question the next 12 months will answer empirically.
The Counterargument — Taken Seriously
The strongest version of the case for EF’s restructuring is not “leaner is always better.” It is something more specific: the Ethereum Foundation’s large, centralized research operation was producing work that was not clearly connected to Ethereum’s most urgent competitive needs, and the budget structure was not sustainable regardless of whether the work was good.
PSE’s research was real and technically impressive. MACI and Semaphore are used by a real, if small, set of applications. But Ethereum’s competitive pressure in 2026 is not primarily about L1 privacy. It is about transaction throughput, cost, and developer experience — areas where Solana has demonstrably closed the gap and in some respects exceeded Ethereum’s user-facing performance. A ZK privacy research lab is a long-horizon investment in capabilities that may matter significantly in five years and are essentially invisible to the retail users and application developers determining market share today.
The endowment model is a bet on durability over intensity. An EF that cannot run out of money in any foreseeable scenario — because it is only spending returns, not capital — is structurally more resilient than one optimizing for maximum research output per year at the cost of a finite runway. The L2 teams that do the most user-facing development have independent resources. The infrastructure that Ethereum restaking and EigenLayer represents is funded and governed independently of the EF. The Ethereum protocol does not require a large EF to function, even if it requires a functional one.
The Bitcoin development parallel also holds up better than critics acknowledge. Bitcoin Core’s key protocol upgrades — Taproot, SegWit, CLTV and CSV time-lock changes — were all coordinated without a large centralized foundation. They took time and required broad social consensus. But they shipped. The argument that Ethereum’s higher upgrade cadence requires centralized EF coordination assumes a development model that may itself be due for reassessment at Ethereum’s current maturity and scale.
What to Watch Over the Next 12 Months
The specific questions that will determine whether the EF’s restructuring represents a controlled transition or a capacity loss have answers that will become visible on a definable timeline.
Who carries the PSE work forward will be visible within two quarters. If MACI, Semaphore, and the “prove anywhere” ZK research get picked up by an independent team with adequate funding — through Ethlabs, through an L2 team, through a new grant-funded organization — the applied cryptography gap PSE’s closure created is filled. If it is not picked up, the CROPS “privacy” mandate becomes aspirational, and L1 privacy becomes a goal without an organization executing it.
Whether the All Core Devs process maintains velocity through the leadership transition is testable by the end of 2026. The EF coordinates ACD calls, manages EIP repository governance, and provides the institutional continuity for upgrade coordination. If the next major upgrade cycle shows slippage against expected timelines, the coordination risk the departures created will be visible in the on-chain record. If timelines hold, the concern was overstated.
Whether the endowment math holds depends on ETH performance over the next 24 months. If Ethereum’s institutional flow picture improves — perhaps through staking yield becoming available through SEC-approved ETF structures — the treasury grows and the 5 percent spend rate buys more research capacity. If ETH continues to underperform BTC on the ETF flow metrics that have characterized June 2026, the endowment model will produce a smaller EF in real-dollar terms than the current announcement implies. The EF bet on sustainability. Sustainability in an endowment model depends on what you are endowed with — and the EF is endowed with ETH.
- The Defiant — Ethereum Foundation Cuts 20% of Staff in Sweeping Reset (June 23, 2026)
- TechTimes — Ethereum Foundation Cuts 54 Jobs, Shuts ZK Research Lab, Slashes Budget 40% (June 23, 2026)
- The Defiant — Five Former EF Researchers Launch Ethlabs Independent R&D Lab
- CoinDesk — Ethereum Foundation Cuts 20% of Staff Amid Leadership Exodus (June 23, 2026)
- CoinDesk — Hsiao-Wei Wang Resigns as Ethereum Foundation Co-Executive Director (June 18, 2026)
- Startup Fortune — Five Former Ethereum Foundation Researchers Launch Ethlabs
- Bitcoin.com News — IBIT Leads $86M Bitcoin ETF Inflow as Ethereum Funds Extend Outflow Streak (June 2026)
- Thirdweb Blog — Ethereum Foundation Restructuring: 40% Budget Cut and the Ethlabs R&D Lab Built by EF Alumni
