AI’s Role in Blockchain

BLOCKCHAIN'S NEXT LEGAL QUESTION ISN'T CRYPTO, IT'S AI

For years, the central legal question in digital assets has been some version of the same thing, is this token a security. That question has generated a substantial body of guidance, enforcement history, and industry practice built around the Howey test and its "efforts of others" prong. A newer development is starting to complicate that analysis in a way the existing framework was not built to anticipate, the "others" doing the efforts is increasingly an AI system, not a person.

This article is general legal education for founders, fund counsel, family offices, and advisors working in digital assets. It is not advice about any specific token, protocol, or company, and nothing here should be treated as legal advice for your situation. Every project's facts are different, and securities law analysis in this space is highly fact dependent, which is why anyone evaluating a specific token or structure should consult qualified counsel directly.

WHY THE HOWEY TEST WAS BUILT FOR HUMAN EFFORT

The Howey test asks, among other things, whether an investment involves an expectation of profit derived from the efforts of others. That framework assumes, reasonably for most of its history, that the "others" are people, a founding team, a promoter, a manager, someone whose ongoing managerial effort is what an investor is relying on for returns.

A significant amount of digital asset legal analysis over the last several years has focused on exactly this question, how much ongoing effort by an identifiable team is baked into a token's value proposition. Tokens with minimal ongoing managerial effort, where the network is sufficiently decentralized and functional on its own, have generally had a stronger argument against being classified as securities. Tokens where a central team's continued work is clearly what drives value have faced a harder version of that argument.

WHAT CHANGES WHEN AN AI SYSTEM IS DOING THE WORK

Increasingly, protocols and platforms are deploying AI systems to perform functions that used to require a human team, automated yield strategies, algorithmic trading, treasury rebalancing, liquidity management, and similar ongoing, active functions tied directly to a token's value or return profile.

This raises a question that does not have settled precedent yet, if an AI system is doing that active, managerial style work instead of a human team, does that strengthen or weaken an argument that the token involves efforts of others under Howey. There are reasonable arguments on more than one side of this.

One view is that an AI system performing these functions is still traceable back to a person or entity, the developer who built it, deployed it, or maintains it, and that the efforts prong should still apply through that chain of responsibility, arguably even more clearly than in a diffuse human team scenario. Under this view, automating the function does not remove the human effort, it relocates it earlier in the process, into the design and deployment of the system rather than its ongoing operation.

A different view is that if the AI system operates with genuine autonomy, without ongoing human intervention, the case for classifying it as continued "efforts of others" becomes less clear, since no person is actively managing the outcome day to day. This view is less tested and more likely to face regulatory skepticism, since regulators have generally been cautious about arguments that automation removes accountability rather than simply changing its form.

Neither view has been definitively resolved by courts or regulators as applied to this fact pattern, which is precisely why this is a developing area rather than settled law.

WHY THIS MATTERS BEFORE A TOKEN IS STRUCTURED OR LAUNCHED

For founders and protocols building AI-driven token functions, this is not a theoretical question to revisit later. The way an AI system's role is designed, disclosed, and documented at the outset can materially affect how a securities analysis plays out down the line. A token whose value proposition is described publicly as "managed by an autonomous AI system" invites a different level of regulatory scrutiny than one where the same function exists but is not marketed around ongoing managerial activity, even if the underlying mechanics are similar.

This also matters for fund counsel and family offices evaluating digital asset allocations. A token backed by an AI-managed yield or trading function may carry a different risk profile from a securities law standpoint than a comparable token without that feature, even if both are marketed similarly to investors. Diligence in this space increasingly needs to ask not just what the token does, but what is doing it, and how that activity is described to the market.

WHERE THIS IS HEADED

Regulatory guidance specific to AI managed digital asset functions is still developing, and it would not be accurate to say this question has a clear, settled answer. What can be said is that the intersection of autonomous systems and the efforts of others analysis is an active area, and structures built without considering it are more likely to face a harder path if scrutinized later. Involving counsel during the design phase, rather than after a token has launched, tends to be the more defensible approach.

FREQUENTLY ASKED QUESTIONS

1. Does using an AI system automatically make a token a security.

No. There is no automatic rule. Whether a token is a security depends on the full Howey analysis, including whether there is an investment of money, a common enterprise, and an expectation of profit derived from the efforts of others, considered against the specific facts of the token and how it is marketed.

2. How does the Howey test apply to AI managed crypto assets.

The Howey test itself has not been rewritten for AI. The open question is how the "efforts of others" prong applies when an AI system, rather than a human team, is performing the ongoing managerial functions tied to a token's value.

3. Can an AI agent be considered the "other" whose efforts investors are relying on.

This is an unresolved question. Some analyses trace responsibility back to the developer or operator of the AI system, while others focus on whether the system operates with genuine autonomy. Neither approach has been definitively resolved by courts or regulators as applied to this fact pattern.

4. Does decentralization still matter if an AI system is running the protocol.

Yes, decentralization remains a relevant factor, but the presence of an AI system performing active management functions may itself be evidence of ongoing managerial effort, depending on how it is designed and disclosed.

5. What should founders document when deploying an AI managed token function.

Generally, how the AI system was designed, what decisions it makes autonomously versus under human oversight, and how its role is described publicly. These details can materially affect a later securities analysis and are best addressed with counsel before launch.

6. Are regulators actively looking at AI and digital assets together.

This is a developing area, and regulatory attention to the intersection of AI and digital assets has been increasing, though comprehensive, settled guidance specific to this exact question does not yet exist.

7. Does marketing a token as "AI managed" increase legal risk.

It can, since public descriptions of ongoing active management, whether performed by a person or a system, are relevant to the efforts of others analysis. How a token's function is marketed is a factor worth reviewing with counsel.

8. Should fund counsel or family offices evaluate this differently in diligence.

Many are beginning to, since a token supported by active AI managed functions may present a different risk profile from a securities law standpoint than a comparable token without that feature, even if the two appear similar on the surface.

9. Is this issue specific to any one type of digital asset.

No. It is potentially relevant to any token or protocol where an AI system performs ongoing functions tied to value or returns, including yield generation, trading, treasury management, or liquidity provisioning.

10. When should a blockchain project involve legal counsel on this issue.

Generally during the design phase, before a token is structured, marketed, or launched, since decisions made at that stage are more difficult to revisit once the project is public and operating.

This article is provided for general informational purposes only and does not constitute legal, financial, or investment advice. It does not create an attorney-client relationship between the reader and Kaelus Law, PLLC. Every project's circumstances are different, and readers should consult qualified legal counsel before making decisions regarding token structure, securities compliance, or digital asset offerings. Attorneys at Kaelus Law are licensed to practice law in certain jurisdictions within the United States and this content is not intended to constitute advertising or solicitation in jurisdictions where such content would not comply with applicable rules.

Kaelus Law 

Precision. Alignment. Resolution. 

Contact us at 833-900-7890 or info@kaeluslaw.com for more information.