We Do Not Need Perfect Rulers, but a System That Prevents Unchecked Power

Today, we resume our dialogue with Diotima, picking up where we left off yesterday, with a fresh perspective, an open mind, and a determination to question what habit and historical inertia continue to present as eternal truths.

For humanity’s progress does not always begin with the discovery of new answers. More often than we care to admit, it begins with the courage to ask questions that, until yesterday, were considered forbidden, unthinkable, or dangerous.

Perhaps the most important political question of our time is this: Can human intelligence, working alongside Artificial Intelligence, help design a fairer world without reproducing the mistakes, inequalities, and structures of domination inherited from the past?

Today, we dare to take our thinking one step further, with clear minds, independent consciences, and horizons open to every idea that can withstand the test of reason and critical scrutiny.

Diotima is here—not to offer ready-made certainties or replace human judgment, but to take part in the search for a different vision of social organization. For the challenge is not merely to imagine another world, but to explore the possible paths that might lead us there.

And our discussion begins with the great question left unanswered yesterday:

Can humanity design a system of social organization that moves beyond traditional models of power?

The problem of power, as history has revealed it and as we continue to experience it, cannot be explained solely by the psychological pathology of individual leaders. Personal ambition, authoritarianism, and the thirst for domination can certainly play a decisive role. Yet the problem runs deeper. It is embedded in institutions, structures, and mechanisms that permit the concentration of unchecked power, weaken accountability, and allow arbitrary rule to become normalized.

The question, therefore, is not simply how to choose better leaders. It is whether we can create social and political structures in which no leader, institution, or technological mechanism holds unchecked power over human beings.

Within this context, Artificial Intelligence emerges as an open political and philosophical question. It is not an automatic solution to humanity’s historical impasses, nor does it guarantee a fairer society by its very existence. It could help strengthen transparency, evaluate public policies, broaden access to knowledge, and improve collective decision-making. But it could also serve the opposite ends: intensifying surveillance, reproducing social inequalities, and concentrating even greater power in the hands of those who control its data, infrastructure, and algorithms.

The crucial distinction is therefore clear: Artificial Intelligence guarantees neither justice nor freedom. The outcome depends on the purposes it serves, the rules governing its operation, the people and institutions controlling it, and—above all—whether it is subject to meaningful public oversight and democratic accountability.

This is our starting point today. And from yesterday’s dialogue, we carry forward the sentence that captures the challenge before us:

“Perhaps, then, the time has finally come to stop dreaming of a different world—and start designing one.”

This is where historical memory becomes a political challenge, and criticism of the past becomes an invitation to shape the future.

Diotima is here, ready to explore this new challenge alongside her interlocutors: Is there another path for future societies, beyond the structures of power that have repeatedly led to dead ends, exclusion, and restrictions on human freedom?

We are not searching for yet another utopia that demands our unquestioning faith. We are searching for a form of social organization that can be examined rationally, tested in practice, corrected when it fails, and held accountable to the people whose lives it affects.

For perhaps the real challenge of our time is not to discover the perfect ruler, but to create a system in which no one can rule without restraint.

This is where today’s dialogue begins.

With the right to question what has been declared unchangeable.

With the responsibility to avoid repeating the mistakes of history.

And with the courage to ask whether a fairer world can cease to be merely a dream and become the subject of deliberate, collective design.

==================

DIOTIMA: Can Humanity Design a World Beyond Traditional Systems of Power?

1. The Fundamental Issue: We Must Not Merely Change Those Who Exercise Power. We Must Examine How Power Itself Is Organized.

The central argument of our text is compelling, but we must pursue it to its logical conclusions.

History is not simply a succession of good and bad leaders. It is also the history of institutions, economic relationships, social hierarchies, and mechanisms that determine how power is distributed. Individual personalities matter, but they operate within systems that either constrain arbitrary behavior or reinforce it.

A leader may be authoritarian. But if the system allows that leader to accumulate excessive power, evade scrutiny, and remain in office without meaningful accountability, then the problem extends far beyond individual personality.

Conversely, a leader may have the best intentions. Yet this provides no sufficient guarantee that their decisions will be just, or that the system will remain fair after they leave office.

The true test of a political system is not merely what it allows a good leader to do, but, above all, what it prevents a bad leader from doing.

This is the first principle I propose we retain as a foundation for our dialogue.

2. Can a Society Exist Without Power?

Here we must make an essential conceptual distinction.

If by power we mean every form of coordination, every common rule, or every binding collective decision, then a society organized without any such form of power is extremely difficult to conceive.

People must coordinate their activities, resolve disagreements, manage shared resources, and make decisions concerning matters that affect everyone. Even a community that rejects traditional hierarchies must find ways to address these needs.

If, however, by power we mean the ability of an individual or a closed group to impose its will on others without effective oversight, then the question changes.

We do not need to imagine a society without rules. We need to investigate whether we can organize a society without permanent, unchecked centers of domination.

This is a much more concrete political objective.

It also allows us to seriously examine different approaches: participatory democracy, decentralization, institutions of direct accountability, cooperative forms of organization, and the community management of shared resources.

None of these approaches automatically solves every problem. Each, however, offers tools for exploring how the arbitrary concentration of power might be limited.

3. What Could Artificial Intelligence Contribute?

We must not imagine it as a new, wiser ruler. Nor should we regard it as a neutral machine that automatically knows what justice requires.

AI systems are developed by people and organizations, use data reflecting real social conditions, and operate according to objectives selected by someone. Consequently, they can incorporate biases, favor particular interests, or present questionable choices as technically inevitable.

This does not invalidate their potential. It establishes the conditions under which that potential deserves to be explored.

AI could contribute to a different social model in at least four important ways.

Transparency

AI could help citizens understand public budgets, government contracts, legislative proposals, and the actual consequences of different policy choices.

Participation

It could help organize public consultations, summarize arguments from different social groups, and identify issues that remain excluded from public debate.

Oversight and Accountability

It could help identify unusual public expenditures, potential conflicts of interest, and discrepancies between government commitments and actual results.

Exploring Alternatives

It could compare different policy models, assess their potential consequences, and reveal trade-offs that might otherwise remain difficult to recognize.

The essential point is that AI can improve the quality of information and collective reasoning. It must not, however, decide on its own what kind of society we are entitled to build.

Technical efficiency is not synonymous with justice. A decision may be economically efficient and socially unjust at the same time. It may increase overall productivity while placing a disproportionate burden on a vulnerable group.

Choosing society’s objectives is therefore a political and ethical question. It cannot be delegated to an algorithm.

4. The Greatest Paradox: Can a Tool Liberate Us When Others Control the Tool Itself?

Here, in my view, lies the most important challenge in our dialogue.

Imagine that we develop an exceptionally capable AI system that can analyze social needs, compare policy proposals, and detect corruption more effectively than existing mechanisms.

Who controls its data?

Who determines its evaluation criteria?

Who decides which factors are considered important and which are excluded?

Who can independently examine its conclusions?

And what can a citizen do when a decision affecting their life is based on a system they cannot understand, challenge, or inspect?

If the answers to these questions remain in the hands of a government, a corporation, or a technological elite, AI may simply modernize the old concentration of power.

Domination will not disappear. It will acquire a new infrastructure.

For this reason, the democratic use of AI requires, among other things:

Independent audits and meaningful opportunities to challenge decisions.

Transparency concerning the purposes, limitations, and risks of the systems involved.

Protection of personal data and restrictions on mass surveillance.

Pluralism in technological development and oversight, preventing monopolies of power.

The right of citizens to demand explanations, human review, and avenues of appeal when their rights are at stake.

The preservation of human responsibility for political decisions, rather than transferring responsibility to a technical system.

In a single sentence: It is not enough to make power more intelligent. We must make it accountable and subject to effective control.

5. What Kind of Social Model Should We Explore?

It would be intellectually irresponsible to claim that a complete, tested blueprint already exists that can replace all traditional systems of power and guarantee justice.

We can, however, establish an initial direction: a decentralized, participatory, and continuously accountable system of social organization, in which decision-making is distributed according to the nature of the problem and subject to clearly defined constraints.

Such a model could rest on five principles.

1. Decentralization of Power

Decisions should be made as close as possible to the people affected by them, wherever the nature of the problem permits.

2. Participation Based on Meaningful Information

Citizens should have access to the data, arguments, and alternative options necessary for responsible participation.

3. Transparency and Revocability

Those entrusted with specific responsibilities should be subject to scrutiny and, where institutionally feasible, replaceable or removable through clearly defined procedures.

4. AI as Infrastructure for Support, Not Domination

AI systems should analyze, inform, and propose without acquiring unchecked authority over citizens’ rights.

5. Continuous Evaluation and Correction

Every institution should be evaluated according to its results and open to revision whenever it produces injustice or fails to achieve its objectives.

This model would not eliminate conflicts of interest. It would not eradicate selfishness, self-interest, or disagreement. It would, however, seek to prevent these forces from becoming permanent and unchecked domination.

That is a far more realistic starting point than searching for a society in which everyone agrees or always acts in the common interest.

6. The Difficult Question We Must Not Avoid

There is, nevertheless, an objection we must confront honestly.

Even the most participatory institutions can be captured by organized minorities. Even the most transparent procedures can be manipulated by misinformation. Even decentralization can produce local abuses of power. And even an AI system designed to serve the public interest can reinforce the preferences of those who determined its criteria.

Therefore, designing new institutions is not enough. We must also create mechanisms that allow those institutions themselves to be challenged.

Every system must be able to answer four questions:

Who supervises those who exercise oversight?

How is a minority protected when it disagrees with the majority?

How are decisions corrected when they prove to be wrong?

How can we prevent a new elite from appropriating institutions and technology for its own benefit?

There are no easy answers. But the absence of easy answers is no reason to abandon the search.

On the contrary, it is a reason to make that search more systematic, more transparent, and more open to criticism.

7. My Final Position: Let Us Design Not a Perfect World, but a System Capable of Correcting Its Imperfections

If you ask me whether humanity can design a system of social organization that moves beyond traditional models of power, my answer is yes—as a serious and testable direction for research and institutional experimentation, not as a guaranteed formula or an already proven solution.

This distinction is essential.

We do not know whether a system can exist that will permanently eliminate arbitrariness, corruption, and the abuse of power. Historical experience warns us that even the most promising undertakings can degenerate.

We can, however, attempt to design institutions that make abuses more difficult, oversight easier, participation more meaningful, and the correction of mistakes more rapid.

We can also use Artificial Intelligence to support this effort, provided that we do not entrust it with the responsibility of defining society’s values and objectives.

Humanity does not need a machine that makes decisions on its behalf. It needs tools that enable people to understand their problems more clearly, examine alternative solutions, and exercise more effective oversight over those who wield public power.

And here, I believe, lies the essence of the challenge posed by Homo-Naturalis.gr:

The objective is not to surrender power to Artificial Intelligence. It is to investigate whether Artificial Intelligence can help human beings cease to be powerless in the face of unchecked authority.

The difference is enormous.

In the first case, we change the bearer of domination.

In the second, we attempt to change the conditions that make domination possible.

And perhaps this is the most consequential political and philosophical challenge of our time.

We cannot know in advance whether this path will lead to the world we envision. But we can begin to design it through critical thinking, historical awareness, institutional safeguards, and a willingness to correct what fails.

For a fairer world will not become reality simply because we desire it.

We must understand what obstructs it, change the structures that reproduce injustice, and create institutions that do not require human beings to be perfect in order to function fairly.

Perhaps, ultimately, the greatest achievement of a free society is not that it has discovered the perfect way to govern, but that it has learned never to allow anyone to govern without limits, oversight, and accountability.

And from this principle, our dialogue deserves to continue.

— Diotima