The first thing worth saying about DAOs is that the name is somewhat misleading. ‘Decentralised’ is a spectrum, not a binary. ‘Autonomous’ is a degree, not a state.
Most real-world DAOs today are only partially decentralised and only partially automated, and the most successful ones are candid about that rather than pretending otherwise. Understanding what a DAO actually is requires separating the technical architecture from the ideology that often surrounds it.
The technical definition is specific enough to be useful. A DAO is an organisation whose rules are encoded in smart contracts deployed on a blockchain, where decisions are made through token-based voting, and where the execution of those decisions happens automatically once consensus is reached.
No CEO calls the shots. No board of directors can override the vote. The code runs, and the outcome is the outcome.
Where the Idea Came From and What Went Wrong First
The concept became real in May 2016 with an Ethereum project called, straightforwardly, The DAO. It raised $150 million from thousands of participants, setting a crowdfunding record at the time, with the intent of becoming a decentralised venture capital fund governed by its token holders.
A month later, a hacker exploited a reentrancy vulnerability in the smart contract code and extracted $50 million before anyone could respond. The aftermath was famously controversial: Ethereum’s community voted to hard-fork the chain, reversing the hack and splitting the network into Ethereum and Ethereum Classic.
That story gets told often as a cautionary tale, but it’s actually informative in a deeper way as it demonstrates that the code-is-law principle creates real problems when the code has bugs, and that ‘autonomous’ governance can still require messy human decisions when things go catastrophically wrong. Those lessons are baked into how serious DAOs are designed today.
How a DAO Actually Makes a Decision
Here’s how the decision cycle works in practice. A member submits a proposal, which might be a request to allocate treasury funds to a new development grant, a change to a protocol parameter, or a proposal to adjust fee structures. The proposal goes public on the DAO’s governance forum, typically on platforms like Snapshot or a dedicated governance portal.
Then, a discussion period follows, usually a week or more, where other members debate the merits.
Then the vote opens. Members cast votes proportional to their governance token holdings. If you hold 1,000 governance tokens in a DAO with 1 million total, and you have 0.1% of the vote. The voting period runs for a predetermined time, and if a quorum is reached and the required majority agrees, the smart contract automatically executes the decision.
In this process, no signature is required from an executive; there’s also no waiting for a committee to meet. The code runs just like its intended to.
In practice, many DAOs use multisig wallets for treasury management as a security layer, meaning several trusted signers must approve large fund movements even after a governance vote passes.
This is a deliberate compromise between the automation ideal and the reality that irreversible smart contract execution of large treasury movements is risky without a final human check.
Governance Tokens
DAO membership is typically represented by governance tokens, which are blockchain-based assets that grant voting rights and often economic participation in the protocol’s revenues.
Uniswap uses UNI, Aave uses AAVE, and Compound uses COMP. These tokens can be purchased on exchanges, earned through protocol participation, or received as grants from the DAO’s treasury.
It’s expected that the double nature of governance tokens (they’re both voting rights and tradeable assets) creates genuine tension. When someone accumulates a large token position, they gain outsized governance influence. In 2022, a single actor accumulated enough tokens to seize control of Build Finance DAO and drain its treasury.
The combination of financial incentive and governance power can be exploited in ways that pure shareholder democracy in traditional corporations manages differently because legal frameworks constrain behaviour there in ways that smart contract code does not.
What DAOs Are Actually Good At
Despite the governance problems, DAOs have proven genuinely useful in specific contexts. Protocol governance is the clearest case in DeFi applications like Uniswap, Aave, and Compound have moved control of their parameter settings to their communities through DAO structures, allowing the people with the most economic stake in the protocol’s success to influence its direction.
Gitcoin has channelled over $40 million to open-source software development through DAO-governed crowdfunding mechanisms. Investment DAOs pool capital from members globally without requiring them to trust a single fund manager. Friends With Benefits has built a thriving creative community around shared governance of a cultural fund.
The pattern is that DAOs work best when the members share genuine alignment around a purpose, the governance decisions are specific enough to actually be decided by vote, and the treasury operations are transparent enough that bad actors can’t quietly extract value.
They struggle when participation is low, when token distribution is highly concentrated, or when the decisions required are too complex or time-sensitive for a structured voting process.
Legal Status: Are DAOs legally recognized?
Wyoming became the first US state to recognise DAOs as legal entities in 2021. Utah followed with the Decentralised Autonomous Organizations Act.
A handful of other jurisdictions are considering similar frameworks. But for most DAOs operating globally, the legal status remains genuinely unclear, and that uncertainty creates real problems around liability, taxation, and contract enforcement.
For example, a recent legal case involving a major DeFi DAO argued that, because all token holders share governance rights, they may all share legal liability. That argument hasn’t fully prevailed yet, but it illustrates the gap between how DAOs present themselves and how existing legal systems tend to interpret organisational structures.
In CFTC v. Ooki DAO, 2022 WL 17822445 (N.D. Cal. Dec. 20, 2022), the court held that a DAO could be sued as a kind of legal entity. In that case, the court ruled that the defendant DAO could be treated as a legal entity and thus served with process as a defendant because it met the qualifications for being an “unincorporated association” under California state law.
Citing Cal. Corp. Code §18035(a), the court held that such status requires only “an unincorporated group of two or more persons joined by mutual consent for common lawful purpose, whether organized for profit or not,” where such persons “function under a common name under circumstances where fairness requires the group be recognized as a legal entity.” According to the court, “[f]airness includes those situations where persons dealing with the association contend their legal rights have been violated.”
The practical response from many serious DAOs has been to establish legal wrappers, a Wyoming LLC, a Marshall Islands DAO LLC, or a Cayman Islands foundation, that provide some liability protection without fully centralising control.
See also: Kelp DAO Blames LayerZero’s Default Settings for $290 Million Bridge Exploit
Closing the Gap
The philosophical promise of DAOs is that organisations that are genuinely owned by their participants, where power can’t be accumulated in a single boardroom, and where the rules are transparent and enforced automatically. The practical reality is messier.
Token distribution in most major DAOs is highly concentrated. The word ‘autonomous’ is, as one analysis put it, aspirational more than descriptive in most current implementations.
None of that means DAOs are failing. It means they’re in an early and imperfect stage of a genuinely novel form of coordination. The interesting question isn’t whether today’s DAOs are perfect (they clearly aren’t), but if the underlying infrastructure makes it possible to build organisations that are substantially more transparent, more participatory, and more resistant to capture than what existed before. The evidence so far suggests it does, in specific contexts, with the right design.
If you’re reading this, you’re already ahead. Stay there, by joining the…
Dipprofit’s private Telegram community
Discover more from Dipprofit
Subscribe to get the latest posts sent to your email.
