SEC staff have clarified how shared governance fits into the agency’s crypto framework. The questions reach beyond token votes to the people who can change a network, control access and carry out decisions after a proposal passes.
Summary
- SEC staff added 3 crypto governance answers on November 2.
- Questions 1.4 to 1.6 address governance groups, participation controls and functionality.
- The underlying SEC interpretation was issued on March 17 and became effective on March 23.
- Uniswap’s governance process includes a 7-day vote between a 2-day delay and a 2-day timelock.
- The CFTC’s 2023 Ooki DAO judgment imposed a $643,542 penalty on the organisation.
SEC staff have added 3 answers to their crypto asset guidance explaining how governance participation and control over a network fit into token classification.
The SEC’s Division of Corporation Finance updated its FAQ on November 2, adding questions 1.4 to 1.6. The answers distinguish ordinary participation in shared governance from a coordinated group with effective control.
Their starting point is that unrelated participants do not generally become a controlling group simply by taking part in a governance system. The same principle can apply where governance operates through an organisation, association or foundation.
The more demanding question is who can actually act: admit or exclude users, change upgrade rules, control validators or modify the code. A token vote is one part of that inquiry, alongside the permissions through which decisions are implemented.
The SEC governance update is about control, not every voter’s liability
The new answers concern the definition of a central party within the agency’s crypto classification framework. They do not create a universal rule that voting on a token proposal makes someone responsible for everything the project does.
The guidance remains a staff view, with no legal force and no new obligations. The Commission has neither approved nor disapproved the FAQ’s contents.
As crypto.news previously reported, the three new answers distinguish a functional system from a decentralised one. A network can support the token’s intended use while retaining concentrated operational, economic or voting powers.
The update also addresses delegated voting. A publicly visible distribution of tokens does not, by itself, show how the associated voting authority is concentrated after holders delegate it.
Consider a hypothetical project with 10,000 token-holding addresses. If most voting power is delegated to a small number of related organisations, the address count alone would conceal how decisions are actually made.
The reverse can be true as well. Several unrelated delegates may vote for the same proposal without having a separate agreement to operate as a coordinated bloc.
Those examples identify different factual questions; they do not establish a legal finding about either system. The November answers leave the classification dependent on the facts and circumstances.
The original March interpretive release, numbered 33-11412 and 34-105020, distinguishes crypto assets from transactions or arrangements involving them. A non-security token can be sold subject to an investment contract, depending on the associated promises and managerial efforts.
Governance therefore sits within a larger examination of the asset, the system and the offer. Reading only the voting arrangements can miss the representations an issuer made when people purchased the token.
The release’s five-category taxonomy includes digital commodities, digital collectibles, digital tools, stablecoins and digital securities. The new FAQ expands on particular parts of that framework, rather than issuing an individual classification for every governance token.
Voting power and the power to execute can sit with different people
A proposal can attract a large vote and still depend on a small group to carry it out. The relevant record includes both the vote result and the contracts or people authorised to implement it.
OpenZeppelin’s governance documentation separates voting power, quorum, vote counting and execution. Developers configure those components when building a governance system; the presence of a Governor contract does not make every project’s arrangements identical.
One system may execute a successful proposal automatically after a delay. Another may rely on a multisig to make the change, leaving an additional set of approvals between the community’s decision and the resulting transaction.
Those are different operational arrangements even if both describe themselves as token governed. The governance page should make clear whether a vote is binding, advisory or a condition for a separate authorised actor.
The OpenZeppelin access-control model illustrates the distinction through ownership and roles. A role can authorise a specific function, while a timelock delays operations assigned to it.
An administrator can have authority over one contract without controlling every part of the project. A treasury signer, for example, may be able to release an approved grant while having no permission to upgrade the protocol.
The permissions need to be traced function by function. A single label such as “administrator” cannot establish whether the holder can change access rules, spend funds or merely perform a narrowly defined operational task.
Aave’s governance documentation distinguishes proposition power from voting power and separates protocol emergency guardians from governance emergency guardians. Different mandates exist for pausing protocol activity and cancelling malicious or erroneous proposals.
Its documentation provides a concrete example of why the action matters more than the title. Two bodies called guardians can have different responsibilities, with each role answering a different failure scenario.
The SEC update does not provide a project-specific judgment on those Aave arrangements. They illustrate the public records available when examining who can propose a change, who can approve it and who can stop or execute it.
Delegation can concentrate authority without moving tokens
Delegation allows token holders to assign voting power while retaining ownership of the underlying assets. The resulting voting map can differ substantially from the token ownership map.
OpenZeppelin’s Votes module describes voting units and their delegation to a representative. The representative can pool voting power assigned by several holders and use it in governance decisions.
This is a practical response to participation limits: not every holder follows every technical proposal. The resulting authority still needs to be measured using delegated votes and the relevant snapshot, not simply wallet balances.
ENS’s governance process requires holders to delegate voting rights before voting or proposing, including delegation to themselves. It distinguishes executable proposals from other governance decisions.
That arrangement shows why an undelegated holding and an active vote are not the same measure. A project can have many holders while only a portion of its potential voting weight is available in a particular decision.
Take an illustrative supply of 100 million voting tokens. If 40 million are delegated and a delegate controls 12 million of those votes, that delegate holds 30% of delegated power but 12% of the total supply.
Neither percentage is interchangeable with the share of votes cast on a specific proposal. If only 20 million votes are cast, the same 12 million would represent 60% of participation, assuming all were used.
Each denominator answers a different question about control and participation. Reporting a delegate’s share without naming the denominator can make influence look larger or smaller than the actual voting record supports.
The governance system’s quorum introduces another condition. A proposal can have a large majority of votes cast and still fail because insufficient voting power participated under the applicable rules.
Uniswap’s governance glossary distinguishes delegates, quorum, proposals and timelocks. Those definitions let readers connect the token distribution with the actual route through which a proposal becomes a protocol action.
The important evidence is consequently dated: delegated weight at the relevant block, votes cast, quorum achieved and execution status. A current wallet ranking cannot automatically reconstruct the authority available at an earlier vote.
A working network can still retain powerful administrators
Functionality describes whether the asset can be used for the system’s programmed purpose. Decentralisation concerns who retains control over operation and decisions, so an active product does not answer both questions at once.
A network might process transactions continuously while a designated actor controls a validator list. Another might permit open validation but retain a privileged upgrade key over a central contract.
These are hypothetical configurations, not labels assigned to a named token. They show why transaction activity, software availability and governance permissions describe different characteristics of a system.
Arbitrum’s constitution provides a documented example of explicit administrative authority. It assigns chain-owner privileges to governance and a Security Council, with a 9-of-12 threshold for emergency action.
The constitution includes a requirement for a transparency report after an emergency intervention. Its governance framework therefore recognises exceptional human decisions alongside the network’s ordinary programmatic operation.
An emergency mechanism can be designed to protect users while still constituting a permission which merits scrutiny. Its existence should be described precisely, including who controls it and whether normal delays can be bypassed.
L2BEAT’s rollup assessment treats upgrade permissions and emergency authority as separate risk dimensions from validation and data availability. That framework helps distinguish a network’s technical verification from its remaining trust assumptions.
The comparison does not determine securities status. It supplies an operational description of powers that legal analysis may need to examine within the applicable framework.
Uniswap’s published governance process includes a 2-day waiting period, 7-day vote and 2-day timelock for the specified on-chain route. A successful vote and the passage of the execution delay remain distinct events.
The delay gives participants a defined interval before an approved action is executed through that process. Whether every privileged action must follow it depends on the permissions of the actual contracts involved.
A complete account should therefore identify exceptions as well as the standard path. Permission to pause, cancel or upgrade immediately cannot be inferred from a general timetable covering ordinary proposals.
DUNI separates a legal organisation from its service providers
The SEC’s reference to organisations and foundations addresses a practical issue for DAOs: shared governance can require contracts, tax administration and other work outside a blockchain.
Uniswap Foundation’s DUNI explanation describes a Wyoming decentralised unincorporated nonprofit association intended to give governance legal recognition. Its stated uses include entering contracts, managing funds and protecting participants from liabilities associated with the organisation’s collective actions.
The foundation’s position is that legal organisation and community governance can coexist. It does not describe the legal entity as a substitute for the token voting process.
DUNI’s association agreement identifies its purposes and membership conditions. It links membership to UNI ownership and engagement with the governance protocol, including specified forms of participation.
The document separately defines the treasury and governing principles. Those definitions establish a legal relationship around governance activity rather than making every company serving the ecosystem the same organisation.
The formation proposal distinguishes an association agreement, an administrator agreement and a ministerial-agent agreement. Its on-chain preparation included content hashes linking attestations to the off-chain documents.
That structure gives participants records against which to check authority. A service provider’s contractual function can be compared with the decisions reserved to governance and the instructions approved on-chain.
It also creates questions which cannot be answered from the DUNA label alone. Who may amend an agreement, instruct the provider, terminate the relationship or approve an expenditure depends on the governing documents and actual permissions.
Wyoming’s published statutes provide the state-law framework for decentralised unincorporated nonprofit associations. The legislature’s 2026 amendments addressed matters including conversion, definitions and winding-up requirements.
The existence of a state-law entity does not itself resolve federal token classification. Entity powers and participant obligations arise under a different set of provisions from the SEC’s analysis of assets and investment contracts.
The public documents therefore need to be read together. A statement that a DAO has adopted a legal wrapper identifies its organisational form; it does not establish which actor has effective operational control or whether a particular transaction satisfies federal securities law.
The service contracts show what an administrator can actually do
DUNI’s published agreements provide a more detailed example than the term “legal wrapper” alone. They name particular administrative tasks, compensation, duties and termination arrangements alongside the on-chain governance structure.
The administrator engagement agreements assign financial statement preparation, tax work and specified payment administration to Administrator Services, a Cowrie subsidiary. The disclosed engagement runs through filing of the 2026 returns, expected in the third quarter of 2027.
The documents distinguish the company’s administrative services from David Kerr’s separately authorised role signing tax returns. They include fiduciary obligations for the functions described and a continuation period intended to allow replacement of the administrator.
This is task-specific authority. The ability to prepare accounts or make an authorised tax payment is not the same permission as changing a protocol’s consensus rules or redirecting its treasury without approval.
The ministerial-agent agreement separately records the relationship with the appointed agent. It describes independent performance and duties attached to the financial responsibilities assigned under the agreement.
Separate contracts allow different responsibilities to be assigned without treating every provider as the same governing body. They also create records against which a provider’s actual conduct can be compared.
The foundation’s agreement summary distinguishes ordinary members from administrators or agents authorised for specified tasks. It explains that membership alone does not permit a person to sign contracts or act as the association’s representative.
Membership also does not give an ownership claim on DUNI’s assets or profits under the published summary. Governance participation, legal membership and an economic entitlement are therefore different concepts within that structure.
Consider a hypothetical member who votes for a budget but has no authority to sign the provider’s contract. The vote records participation in the organisation’s decision; the appointment and agreement identify who may bind it afterwards.
An administrator may then receive authority to pay an approved invoice from a designated account. Examining that authority requires the agreement and the authorising decision, not an assumption that every person involved in the vote could make the payment independently.
The structure also leaves a factual question about how instructions work in practice. A narrowly written mandate can be compared with the transactions it produces, including whether an expenditure was within an existing agreement or required a fresh governance decision.
The ministerial contract includes a base fee of $100 a month plus amounts specified in accepted assignments. That amounts to $1,200 for 12 months before any additional assignments, using the agreement’s stated monthly rate.
The figure describes remuneration for the appointed role, not the total cost of DUNI’s administration. Other engagements, tax work and approved expenses have their own terms and should not be silently folded into the same number.
The termination provisions also identify who may end the relationship. DUNI can terminate through a valid governance proposal under the agreement, while the agent’s right to leave without cause requires 60 days’ written notice.
Those clauses show a form of continuing oversight which is separate from a token holder’s ability to cast a vote on a protocol upgrade. The organisation needs a process for replacing the person carrying out its off-chain instructions as well as changing software parameters.
The contract does not give every participant a unilateral right to dismiss the provider. An individual objection and a governance decision meeting the governing requirements are different events.
Those records can show a limit on discretion as well as the existence of discretion. A provider performing a specified accounting task is different from a person able to determine the project’s operating rules unilaterally.
The SEC’s FAQ does not classify either arrangement merely because it has a contract. It asks about control within the applicable framework; the contracts supply evidence about the responsibilities which participants have actually accepted.
The distinction cuts both ways. Creating an association does not demonstrate that power is dispersed, and appointing someone to perform a defined service does not demonstrate that the person controls the entire system.
The published terms make the question more concrete: which decision has been delegated, which conditions limit it and which body can replace the provider? The answer may differ between financial administration, protocol upgrades and the allocation of grants.
Past DAO cases asked different legal questions
Token holders have reason to distinguish the SEC classification guidance from litigation over an organisation’s capacity to be sued. Those questions can overlap in the facts without becoming the same legal test.
The Lido dispute began with an investor alleging that LDO was sold as an unregistered security. The original investor lawsuit raised claims about the token’s sale and the role of participants in the organisation.
A November 2024 district-court ruling rejected the argument that the DAO structure alone made the organisation immune from suit. That historical ruling must be distinguished from a final determination of every defendant’s liability.
The court’s November 18 order, in case 23-cv-06492-VC, addressed motions to dismiss. It found the plaintiff had adequately alleged a general partnership and allowed claims against certain institutional participants to proceed.
The pleading posture matters because the court was assessing whether alleged facts could support a claim. It was not conducting a trial establishing that every LDO holder was a partner responsible for every action of the protocol.
Lido governance’s proposal to appoint litigation counsel illustrates another practical problem. It proposed funding an entity to make a limited appearance while expressly disputing that the DAO was a legally cognisable partnership or association.
An organisation can therefore need a mechanism to respond to a lawsuit while contesting the legal theory under which it has been sued. Appointing a representative and conceding the underlying claim are not equivalent actions.
Ooki DAO followed a different path. After the missed response deadlines, the CFTC pursued a default judgment against the organisation.
The agency’s June 2023 statement described a judgment imposing a $643,542 civil penalty, trading and registration bans, and website-related relief. The court held the DAO could be treated as a person under the Commodity Exchange Act.
That result concerned the organisation and the conduct charged in the case. It should not be presented as a blanket judgment that every person voting in any DAO incurs the same obligations.
The November 2 staff answers do not purport to vacate those cases or rewrite state partnership law. Their stated subject is the SEC framework’s classification of crypto systems and the role of a central party within it.
Proposed fundraising rules address another part of the problem
The SEC’s November staff update sits alongside a separate rulemaking process. Its August 18 Regulation Crypto Assets proposal concerns offerings involving certain crypto investment contracts.
The proposal would create two registration exemptions with crypto-specific disclosure requirements and a conditional investment-contract safe harbour. Those provisions were proposed rules, not new exemptions created by the November FAQ.
The fundraising framework included a startup route up to $5 million over four years and a separate fundraising route up to $75 million in a rolling 12-month period. The two thresholds govern different proposed pathways.
A project cannot infer that it qualifies for an offering exemption merely because its token holders vote independently. Eligibility depends on the conditions of the relevant rule, including disclosures and other limitations if the proposal is adopted.
SEC Chair Paul Atkins’ statement on the proposal described the effort as adapting capital-raising rules to non-security crypto assets sold subject to investment contracts. The distinction preserves a separate inquiry into the arrangements surrounding a sale.
Commissioner Mark Uyeda’s statement addressed the difficulty issuers had faced using registration processes designed for other kinds of securities. Both statements explain the agency’s policy rationale; neither replaces the proposal’s operative conditions.
The proposing release, numbered 33-11434, supplies the detailed framework for the contemplated exemptions and safe harbour. It is a different document from the March interpretation and the later staff FAQ.
The three documents therefore occupy different procedural positions. The March release states the Commission’s interpretation, the August document proposes rules, and the November FAQ supplies staff answers to questions arising under the interpretation.
Combining them into a claim that the SEC has granted every DAO a new exemption would lose the status and scope of each action. The relevant question is which document applies to the asset, offer or governance arrangement being examined.
The same discipline applies to legislation. A staff answer does not complete a congressional vote, establish a statutory registration category or substitute for conditions that legislators may ultimately enact.
The documents that reveal control are already public in many projects
A useful assessment begins with the contracts and agreements assigning authority, then compares them with actual decisions. A project’s description of itself can be checked against who proposed, approved, cancelled and executed its latest material changes.
The governance contract identifies voting rules and proposal status. Access-control records identify privileged roles. Treasury transactions identify who released assets and where they went.
Service agreements add another part of the record: which decisions providers may make independently and which require community approval. DUNI’s separate agreements illustrate why an administrator’s legal function should be distinguished from governance itself.
For delegated systems, influence should be measured at the relevant snapshot. A delegate’s current weight can differ from the weight available when an earlier proposal passed.
Aave’s voting explanation describes voting power through eligible balances and delegations on Ethereum. Its proposal guide separates approval requirements from the cross-chain process used to execute a successful decision.
The gap between voting and execution can therefore involve both time and infrastructure. A proposal may pass on one chain while messages still need to reach the contracts on another.
The records can also show limits on a powerful actor. A role that can cancel a malicious proposal may be unable to spend the treasury, while a treasury administrator may be unable to change protocol software.
Those boundaries are concrete facts a project can publish and a reader can inspect. They support a more precise account than a count of token holders or a claim that all governance occurs “on-chain”.
The November FAQ leaves individual classification dependent on the circumstances. It offers no list declaring named governance tokens approved, no universal liability exemption and no project-specific finding that a particular set of emergency powers is acceptable.
Its footnotes define the features staff associate with broad-based governance: timely public records, pre-established procedures and participation not concentrated through affiliated or coordinated stakeholders, including delegation.
What to watch
- Permission changes: Transfers or removal of upgrade, access-control and emergency roles.
- Delegated concentration: Voting weight at proposal snapshots, with related entities and coordinated arrangements identified where documented.
- Execution records: Whether approved proposals take effect through the announced process and waiting periods.
- Entity agreements: Which decisions a foundation, association, administrator or service provider may make independently.
- Rulemaking status: Further Commission action on Regulation Crypto Assets, separately from staff FAQ updates.
FAQ
What did SEC staff change on November 2?
They added questions 1.4 to 1.6 on governance groups, participation controls and the distinction between functionality and decentralisation.
Does voting automatically make someone a central party?
No. The guidance distinguishes ordinary participation by unrelated holders from coordinated arrangements and effective control, with the facts determining the outcome.
Can a foundation coexist with decentralised governance?
The staff answers allow for that possibility. The organisation’s form alone does not establish who controls the system.
Is a functional network necessarily decentralised?
No. A token can perform its programmed role while operational, economic or voting authority remains concentrated.
Did the SEC approve Uniswap or another governance token?
The FAQ supplies general staff answers. It does not grant individual approval or a universal exemption to named tokens.
Do the answers eliminate DAO participant liability?
They address classification within the SEC framework. State-law entity questions, particular conduct and court claims require separate analysis.
Is Regulation Crypto Assets already available as a new exemption?
The August document discussed here is a proposal with separate conditions and procedures. The November FAQ did not adopt it.
What records help establish who controls a project?
Relevant records include role permissions, delegated votes, governing agreements and executed proposal transactions. This is educational analysis, not investment advice.
