There is a particular kind of quiet that settles over a policy document when it is destined to die. It is not the silence of a completed task, nor the hum of a machine at work. It is the silence of a draft being passed from desk to desk, read, annotated, and then carefully, deliberately, set aside. Over the past several weeks, I have been tracking the fate of the Trump administration's proposed executive order on artificial intelligence—the one designed to establish a self-regulatory body for the industry. The order has not been killed, nor has it been signed. It is simply... stalled. For those of us who have spent years watching the intersection of technology and governance, this stall is not a bureaucratic hiccup. It is a loud, unambiguous signal about the deep fractures in how America intends to govern its most transformative technology. And for the blockchain community, which has long championed the idea that code can replace trust in institutions, this moment offers a sobering lesson: self-regulation is only as strong as the integrity of the actors who volunteer for it.
To understand why this executive order is stuck, we must first understand what it represents. The proposed framework is a radical departure from the previous administration's approach. The Biden-era executive order of October 2023 was a sprawling, federal-led directive that mandated safety assessments, required reporting from major AI developers, and coordinated efforts across multiple government agencies. It was, for better or worse, a top-down attempt to impose order on a chaotic frontier. The Trump administration's vision, by contrast, is built on a philosophy of industry self-governance. The proposed order would create a Self-Regulatory Organization (SRO)—a body composed of AI companies themselves, granted federal authorization to police their own ranks. This model has precedent in the financial sector, where organizations like FINRA oversee broker-dealers. But transplanting that model to AI is not a simple administrative change; it is a philosophical bet that the industry can be trusted to regulate itself in the public interest.
Based on my experience auditing whitepapers during the 2017 ICO boom, I can tell you that the gap between a stated commitment to ethics and the actual incentive structure of a company is often a chasm. The SRO model assumes that AI companies will act as responsible stewards of a shared resource. But the history of self-regulation in other industries—from banking to social media—suggests that when profit motives collide with public safety, the public usually loses. The stall of this executive order is not just a political problem; it is a philosophical indictment of the idea that the fox can be trusted to guard the henhouse, even if the fox signs a pledge to do so.
The reasons for the stall are multifaceted, and the signals I have gathered point to a confluence of resistance. The first and most significant obstacle is internal discord within the White House itself. Sources indicate that the draft has been circulated but has not progressed, which typically means that the policy team and the legal counsel are at an impasse. The SRO model requires congressional authorization to be legally binding; an executive order alone cannot grant a private entity the power to regulate its peers. This creates a constitutional vulnerability that the legal team is likely flagging. Furthermore, there is a visible tug-of-war between the national security apparatus, which favors strict export controls and foreign investment scrutiny, and the commerce and technology policy offices, which are pushing for a lighter touch to spur innovation. In an election year, the political calculus is also daunting. Advancing a controversial regulatory architecture that could be perceived as either a giveaway to Big Tech or a betrayal of free-market principles is a risk that a campaign team is unlikely to embrace.
The second source of resistance comes from the very industry the order is meant to empower. On the surface, tech giants support self-regulation because it is far less burdensome than federal mandates. But a deeper analysis reveals a more complex picture. A self-regulatory body dominated by a few large players could be interpreted as a legalized cartel, inviting antitrust scrutiny. Smaller companies, which are often the source of genuine innovation, would be forced to comply with standards set by their larger competitors, potentially raising their costs and creating barriers to entry. And there is a liability concern: if a company participates in the SRO and a member's AI system causes harm, the SRO members could be held partially responsible for failing to prevent it. This is a legal exposure that many corporate counsels are unwilling to accept. The industry's public support for self-regulation is, in many cases, a thin veneer over a deep well of anxiety about the unintended consequences of being asked to police themselves.
The third layer of resistance comes from the states. The stall of the federal order is not occurring in a vacuum; it is happening while a wave of state-level legislation is crashing forward. California has passed SB 53, which mandates safety testing and transparency reports for large AI models, with an expected effective date in 2026. Colorado has enacted the nation's first comprehensive AI consumer protection law, SB 205, which addresses algorithmic discrimination. New York City has already implemented Local Law 144, requiring audits of AI-driven hiring tools. At least forty states have introduced AI-related bills. The federal vacuum is not a void; it is a space that is being rapidly filled by a patchwork of local rules. This fragmentation is a nightmare for companies that operate nationally, but it is also a powerful political force. States do not want to cede their newly claimed authority to a federal SRO that might preempt their stricter rules. The proposed executive order reportedly contains provisions to limit states from creating conflicting regulations, which is a direct assault on federalism that would almost certainly trigger legal challenges. The stall, in this context, is a recognition that the federal government cannot simply override the states without a brutal and costly fight.
The core insight here is that the stall of the AI executive order is not a failure of policy; it is a failure of the foundational premise that self-regulation can function without a clear, enforceable, and universally accepted definition of what constitutes harm. In the blockchain world, we talk about the wisdom of the crowd and the power of decentralized consensus. But those mechanisms work because they are built on transparent, verifiable rules. A self-regulatory body for AI, as currently conceived, lacks that transparency. It would be a closed-door institution where the most powerful players negotiate standards that serve their own interests, all while claiming to act in the public good. This is not a recipe for trust; it is a recipe for a legitimacy crisis.
Let me offer a contrarian perspective, one that I have developed through years of watching both the promise and the peril of decentralized systems. Perhaps the stall is not a problem to be solved, but a feature of a system that is wisely hesitating. The idea that the federal government must act quickly to regulate AI is based on the assumption that inaction is dangerous. But what if the opposite is true? What if the most dangerous outcome is a hastily constructed, poorly designed regulatory framework that locks in bad incentives for decades? The EU AI Act, which came into force in August 2024, is a comprehensive and ambitious piece of legislation. But it is also a rigid, top-down framework that may struggle to adapt to the rapid evolution of AI technology. The United States, by failing to act, is preserving its flexibility. It is allowing the industry to mature, allowing the states to experiment, and allowing the global conversation to develop before committing to a single path. This is a high-risk, high-reward strategy. It could lead to a more thoughtful, more effective regulatory regime in the long run. Or it could lead to a catastrophic safety failure that forces a panicked, overcorrecting response. The stall is a gamble, and the stakes could not be higher.
This brings me to the global dimension, which is where the blockchain community's interest should be most acute. The United States' absence from the global AI governance table is not a neutral event. It is an active transfer of power. The EU AI Act is already creating a 'Brussels Effect,' where global companies choose to comply with EU standards to avoid the complexity of multiple regulatory regimes. This is precisely what happened with the General Data Protection Regulation (GDPR), which became the de facto global standard for data privacy. If the US federal government remains silent for another 12 to 18 months, the EU's AI standards will likely become the baseline for the entire world. This is not necessarily a bad outcome—the EU has a strong track record on consumer protection—but it is a profound loss of American influence. For a nation that has historically led the world in technological innovation, ceding the rule-making authority for its most important new technology is a strategic error of immense proportions. The stall is not just a domestic policy issue; it is a geopolitical retreat.
In my work as an open-source evangelist, I have often argued that transparency is the new currency. This principle applies as much to governance as it does to code. The reason the blockchain community has been able to build trust in decentralized systems is that every transaction, every rule, and every change is visible and auditable. The proposed AI SRO, by contrast, would be an opaque institution. Its decision-making processes would be hidden behind corporate confidentiality. Its standards would be developed in private meetings. Its enforcement actions would be subject to industry capture. This is the antithesis of the transparency that builds trust. The stall of the executive order is, in a sense, a victory for transparency—a refusal to enshrine opacity as the law of the land. But it is a hollow victory if it is not followed by a more honest, more open, and more inclusive process for developing AI governance.
I am reminded of the DeFi Trust Repair Workshops I organized in 2020, in the wake of the bZx hacks. We taught thousands of users how to interact with smart contracts safely, not by relying on the promises of the developers, but by teaching them to read the code themselves. We empowered them to verify, to audit, and to understand the risks. That is the model we need for AI governance. We do not need a self-regulatory body that acts as a black box. We need a framework that empowers the public to understand how AI systems work, to audit their behavior, and to hold their creators accountable. This requires a different kind of regulation—one that focuses on transparency and verifiability rather than on top-down control or industry self-policing.
The signals I am tracking are clear. The executive order is stalled, and the question is whether it will be revived after the election. If it is revived quickly, we will know that the stall was a political strategy, a way to avoid a contentious issue during the campaign. If it remains dormant, we will know that the internal resistance is real and that the SRO model is effectively dead. In the meantime, the states are moving, the EU is moving, and the industry is moving. The only entity that is not moving is the federal government. This is a dangerous moment, not because of what is happening, but because of what is not happening. The absence of federal leadership is creating a vacuum that will be filled by whoever has the will and the power to act.
For the blockchain community, this moment is a call to action. We have spent years building systems that are transparent, auditable, and resistant to capture. We have the tools and the philosophy to contribute to the AI governance debate. We should not be passive observers. We should be active participants, advocating for frameworks that prioritize verifiability over trust, and transparency over opacity. The stall of the AI executive order is an opportunity to push for a better model—one that is built on the principles of open source, decentralized accountability, and human-centric design. Humanity is the ultimate protocol, and it is time we started coding for it.
The risk of inaction is not just a regulatory vacuum. It is the risk of a future where the most powerful technology in human history is governed by a handful of unaccountable corporations, operating behind closed doors, with no obligation to the public good. That is not a future I want to build. And it is not a future that the blockchain community, with its commitment to decentralization and integrity, should accept. We have a choice. We can let the stall become a permanent state of paralysis, or we can use it as a catalyst for a new, more transparent, and more equitable approach to AI governance. The choice is ours, and the time to act is now.
As I look at the stalled executive order, I see not a failure, but a challenge. It is a challenge to the industry to prove that it can be trusted. It is a challenge to the states to coordinate their efforts. It is a challenge to the global community to step up and fill the leadership void. And it is a challenge to all of us who believe in the power of technology to serve humanity, to demand better. The silence in the White House is not an ending. It is a beginning. The question is whether we have the wisdom and the courage to write the next chapter. Building bridges where code ends and trust begins is not just a slogan; it is the work that lies ahead. Auditing ethics before auditing assets is not just a principle; it is the only way forward. Restoring faith in decentralized promises is not just a hope; it is a necessity. The stall is over. The real work is just beginning.