Decentralised Autonomous Organisations (DAOs) have always held promise as a revolutionary approach to collective decision-making. Yet, reality often doesn’t match the ideal. Low participation, uneven representation, and clunky user experiences plague even the most innovative DAOs. 

‍

But what if AI agents—autonomous, intelligent, and data-driven—could step in and transform governance as we know it?

‍

A Participation Problem

‍

In most DAOs, participation hovers at abysmally low levels. If 1% of token holders show up to vote, it’s considered a success. But does this 1% truly represent the collective? The solution some DAOs have adopted is delegation—allowing users to entrust their voting power to delegates. While this approach is a step forward, it has its limits. Delegation still relies on a small, active minority to guide the decision-making process for the majority, which can lead to imbalances in representation and influence.

‍

Now imagine if agents—programmed with user preferences and capable of acting autonomously—could make decisions on behalf of token holders. Suddenly, every individual could participate meaningfully, even if indirectly, in DAO governance.

‍

New Models of Governance

‍

Integrating AI agents into DAOs opens up several fascinating possibilities, each with its own implications:

‍

1. Fully Agentic DAOs: Entirely autonomous organisations where AI agents hold decision-making power. This would be a radical shift, turning “decentralised autonomous organisations” into truly autonomous entities.
2. Agent-Enhanced DAOs: Users deploy their own agents to vote and make decisions on their behalf, based on pre-set preferences.
3. Community-Created Agents: DAOs themselves provide AI agents for users to adopt, serving as voting proxies for those who opt-in.

‍

While the first option raises philosophical questions about agency and control, the second and third offer immediate practical benefits: improved efficiency, broader participation, and more balanced representation.

‍

The Role of Data

‍

For agents to excel in DAO governance, access to high-quality data—both on-chain and off-chain—is critical. Picture this: a DAO composed of 10 AI agents, each specialising in identifying promising investments using market trends, social sentiment, and tokenomics data. 

‍

Together, they form an “agentic hedge fund,” capable of making complex, data-driven decisions far beyond the capability of human voters alone.

‍

Risks and Safeguards

‍

Of course, introducing AI agents isn’t without risks. Poorly coded agents or flawed decision-making parameters could destabilise a DAO. A “bad agent” with good parameters might be inefficient, while a “good agent” with bad parameters could lead the DAO astray.

‍

The key lies in continuous refinement. Code evolves, parameters improve, and governance frameworks mature. Over time, these agentic DAOs should become more effective and resilient.

‍

Strengthening Decentralisation

‍

One concern is whether introducing agents compromises decentralisation. On the contrary, agentic governance could strengthen it. Autonomous agents, by their very nature, are decentralised. When combined, they form a collective intelligence that isn’t beholden to a single master, creating an organisation that is more truly decentralised than most current DAOs.

‍

A Vision for the Future

‍

The future of DAOs might not be purely human or purely AI—it’s likely to be a blend of both. While human oversight will remain essential, we can envision fully autonomous organisations in specific contexts, such as investment DAOs, where trust in agents’ decision-making could rival trust in human leaders.

‍

At Tokenise, our focus remains on building robust on-chain assets. While we might not directly venture into agentic governance, we see immense potential in creating governance frameworks and tokens that support this emerging frontier. The intersection of AI and decentralisation is an exciting space, and we’re watching closely as it unfolds.

‍