Why artificial intelligence must become a freedom technology rather than one more instrument of centralized control
By Félix Maradiaga, president of the World Liberty Congress
October opened in Nashville with two gatherings of global reach, both of great consequence for those of us who work where technology, human rights and the defense of freedom meet. The first was the Global Free Speech Summit, the annual meeting that The Future of Free Speech, the think tank led by Jacob Mchangama, hosts with Vanderbilt University in partnership with Dialogue Vanderbilt. Days later the same city, better known for its music, welcomed Imagine IF, the summit created by Rod Roudi, co-founder of Bitcoin Park. I want to write here about Imagine IF, and about one talk in particular that I consider among the most relevant to the mission of the World Liberty Congress (WLC), a global alliance of pro-democracy leaders and activists from autocratic countries.
At Imagine IF, technology appeared less as a set of separate industries than as a convergence of forces already transforming how power is distributed in our societies. One of the premises that ran through the event could be summed up as the convergence thesis: why artificial intelligence, Bitcoin and energy belong on the same stage, and what we should watch for when we think about them together. All three raise, in the end, one fundamental question. Who will control the infrastructure of the future? States and large centralized institutions alone, or free individuals and communities as well?
For those of us who work with human rights defenders living under authoritarian regimes, that question is not theoretical. It is urgent.
Alex Gladstein, chief strategy officer of the Human Rights Foundation (HRF), offered one of the most stimulating answers in a presentation on giving activists genuine “AI superpowers.” His argument deserves an audience well beyond the technology community. It was, at bottom, less a presentation about software than a reflection on power and autonomy, and on the possibility of radically multiplying the capabilities of individuals who face institutions incomparably larger and better funded than they are.
Gladstein began with an observation that anyone who has worked closely with dissidents will recognize at once. Human rights defenders seldom lack imagination, persistence, courage or ambition. Many have the qualities of exceptional entrepreneurs. They build networks under persecution, improvise with minimal resources, communicate under censorship and keep working even when the work puts them at enormous personal risk.
What they usually lack is staff.
A dictator can call on ministries, intelligence services, media outlets, lawyers, researchers, propagandists, diplomats, financial institutions and thousands of public employees. A human rights defender may have little more than a laptop and three trusted colleagues.
Artificial intelligence can begin to narrow that vast asymmetry. An activist equipped with good AI agents can get help with research, translation, graphic design, data analysis, communications, event planning, fundraising materials, reports and presentations, as well as countless administrative tasks. Work that once took weeks, specialized staff or tens of thousands of dollars now increasingly takes hours.
The promise is not to automate courage. Judgment, friendship, leadership, trust and moral conviction resist automation too. What AI can do is take on a growing share of the work that machines do well, and so free human beings to devote more of their time to what only human beings can do. That distinction matters profoundly.
The politics of technological concentration
Gladstein also posed a harder question: who will control artificial intelligence?
Throughout history, states have shown a natural tendency to treat certain strategic technologies as domains that require government oversight. In some cases the reasons are perfectly legitimate. Biology can produce biological weapons, and chemistry chemical ones. Nuclear physics made possible both the peaceful generation of energy and weapons capable of destroying entire cities. No serious defender of freedom argues that anyone should develop technologies capable of catastrophic harm without rules, safeguards or mechanisms of accountability.
But an enormous difference separates the regulation of dangerous applications from the attempt to establish a political monopoly over knowledge and over the tools themselves. Where artificial intelligence is concerned, that difference is decisive.
The idea is spreading that AI is potentially so dangerous that access to the most powerful models should narrow, step by step, to governments or a small number of large, authorized corporations. The premise is familiar: private individuals are not trustworthy enough to wield technologies this powerful, because an artificial intelligence that slipped out of control could one day threaten humanity.
Legitimate questions about safety exist, and we must take them seriously. We cannot, however, allow apocalyptic scenarios to become a justification for technological centralization.
For authoritarian regimes, that reasoning is remarkably convenient. A government that already seeks to control the press, the universities, the financial system, telecommunications and civil society will naturally find appeal in the idea that artificial intelligence is too dangerous for citizens to use on their own.
We could end up in a world where governments command extraordinary computational capabilities while citizens may use only supervised, limited versions of the same technologies. The right name for that condition is technological dependence, not technological safety.
From the standpoint of the classical liberal tradition, the problem should look familiar. Political liberty depends on the dispersal of power and not only on the good intentions of those who govern. Friedrich Hayek insisted that knowledge is dispersed throughout society and that no central authority can ever possess it completely. That insight takes on striking relevance in the age of artificial intelligence.
The question before us is not whether intelligence should govern society, but whether individuals will be able to own tools that strengthen their own intelligence and their independence.
Technology for freedom
This is why the work of the Human Rights Foundation deserves special recognition.
HRF has moved well beyond abstract debate about technology. It has begun introducing activists directly to tools that can strengthen their autonomy. Its work with Bitcoin, encrypted communications, censorship-resistant technologies, decentralized systems and now artificial intelligence marks an evolution of the first importance within the international human rights movement.
Its “Top 15 Freedom Tech Tools of 2026” is an excellent expression of that philosophy. The list ranges from encrypted and mesh messaging to decentralized social media, Bitcoin payments and self-custody, open-source mining hardware and private AI, with Signal, Bitchat, Damus, Primal, the Lightning Network, Muun Wallet, Cake Wallet, Sparrow Wallet, Casa, Bitaxe, Maple AI, PayPerQ, Atomic Chat and Hermes among them.
These tools do not do the same thing. What unites them is an idea: technology should expand the capacity of individuals to communicate, exchange value, organize, think and act without depending permanently on the permission of a centralized authority.
HRF deserves our gratitude for treating this vision seriously and for leading the effort to bring artificial intelligence to the international community of human rights defenders.
From vulnerable activists to capable networks
This vision closely matches the work we are doing at the World Liberty Congress.
Working with leaders such as Leopoldo López, the Venezuelan opposition figure and WLC co-founder, and drawing inspiration from the ideas championed by Gladstein and other innovators in the freedom tech community, we have placed growing emphasis on the concept of Democratic Decentralized Resistance.
The logic is simple. Authoritarian regimes look for centers of gravity. They jail leaders, seize offices, freeze bank accounts, block websites, infiltrate organizations, cut communications and criminalize entire institutions. A movement that depends on a single headquarters, bank account, leader, communications platform or source of information is vulnerable by definition.
Decentralization transforms that vulnerability. Distributed networks are harder to silence, and independent nodes can survive even when others come under attack. Knowledge can circulate without passing through a single point of control. Funds become more resistant to confiscation, and communications can outlast censorship. Movements can even distribute leadership more widely.
Artificial intelligence now adds a new dimension to this architecture of resilience.
As president of the World Liberty Congress, I have committed to helping artificial intelligence become a natural, everyday tool for human rights defenders around the world, just as we are doing with Bitcoin and other freedom technologies. Activists should not be meeting these technologies for the first time at specialized conferences. They should use them, understand them and experiment with them. In time, they should take part in building them.
From conversation to implementation
Our Freedom Tech Task Force must take on an increasingly important role here.
We should organize AI hackathons centered on the needs of activists, where developers begin by asking what real problems people living under a dictatorship face, not which technology is newest. How can activists document repression more safely? How can a team analyze thousands of pages of government documents in minutes? How can a small organization communicate in several languages at once? How can a movement preserve evidence and evade censorship? How can an organization protect its institutional knowledge when the regime detains its leader?
We must decentralize training too. The WLC can develop a train-the-trainer model, preparing activists in different regions to teach others AI literacy, digital security, decentralized communications and Bitcoin-based tools.
We should also foster the development of secure AI agents and organizational knowledge systems that give small civil society organizations capabilities once reserved for much larger institutions.
But freedom technologies must earn that name. Artificial intelligence makes mistakes. It can invent information, and a poor configuration can expose confidential data. Human oversight must remain at the center, particularly when lives, sensitive sources or strategic decisions are at stake.
The goal is to expand the human capacity to act. Automation for its own sake is no goal at all.
Tyranny must not own the future
I addressed some of these questions on my own panel at Imagine IF, “Imagine IF Tyranny Had Nowhere to Hide,” moderated by Javier El-Hage of the Human Rights Foundation, where I spoke alongside Farida Nabourema of the Togolese Civil League. I will develop those reflections in a separate essay, especially the relationship between new transparency technologies and the vulnerability of authoritarian systems.
Gladstein’s talk deserves a reflection of its own, however, because it captured something essential about this moment in history.
Dictators will use artificial intelligence. That is inevitable, not a remote possibility. What remains open is whether those who resist dictatorships will use it too.
To confront a technologically sophisticated authoritarian state while rejecting the tools of our own time would be, as Gladstein memorably suggested, like fighting a computerized dictator with a typewriter. We cannot allow that.
I also want to congratulate Rod Roudi for creating an exceptional gathering in Imagine IF and for bringing together conversations that too often remain apart: artificial intelligence, Bitcoin, energy, entrepreneurship, human rights and political freedom. In reality, they all belong to a single conversation.
The great political question of our technological era is not whether extraordinarily powerful technologies will exist. They will. The fundamental question is whom they will empower.
The World Liberty Congress will work to ensure that the answer includes dissidents, journalists, students, relatives of political prisoners, independent intellectuals, democratic movements and ordinary citizens fighting to keep their freedom.
Technology must never become exclusively an architecture of control. It can also become an architecture of freedom, and our responsibility is to work to make it so.
Félix Maradiaga is president of the World Liberty Congress.
