The Challenge of AGI and the Destiny of Humanity

What is the real problem with AGI?
It lies in the fact that humanity has not clarified what it seeks from AGI or why it is developing AGI, while fearmongering and narrow obsessions with war and money have taken the lead.
Before anything else, we must make clear why we need AGI, even at the cost of investing enormous resources.

Humanity now stands on the brink of survival.
In the face of this crisis, we must unite and confront it together.
AGI is a powerful means for that purpose.
It is nothing less than the only remaining instrument of salvation.
That is why humanity must develop AGI, no matter the cost.
We must become fully conscious of this necessity.

The True Crisis

What is this existential crisis?
It is the fact that humanity lacks a common foundation for solving problems.
We do have certain shared bases:
mathematics, science and logical positivism, accounting systems, financial systems, the concept of modern law, information and communication technology, network systems, and modern sports.
Yet, we have no established means to integrate them.
The very means of integration is AGI itself.

Today, humanity is required to overcome differences of race, religion, ethnicity, ideology, and political system to solve problems together.
If we fail to cooperate, a catastrophic future is inevitable.
Although it is obvious that only cooperation can avert disaster, divisions and conflicts continue to deepen.
The greatest problem lies in human folly.

The Pressing Global Issues

Humanity faces multiple catastrophic challenges:

  1. Environmental degradation
  2. Overpopulation
  3. War and the management of weapons of mass destruction
  4. Economic instability
  5. Energy scarcity
  6. Food and water shortages
  7. Territorial disputes
  8. Pandemics
  9. Widening inequality

These problems demand immediate solutions. Time is running out.

Against such apocalyptic threats, AGI can provide a shared foundation for solutions.
It is not an exaggeration to call AGI the last hope of humankind.

A Divine Question

Even if we gain the power of God, we cannot surpass God nor become God.
Those who are given power are then tested in how they use it.
If we misuse it, we will be destroyed by that very power.
Indeed, AGI holds the power to annihilate humanity as well as to save it.
But the choice of how to use it belongs to human beings.

Perhaps this is the ultimate question God has given us:
Choose existence or extinction, and choose it yourself.

If you would question the ethics of AI,
first question your own ethics.

God does not speak.
God only shows.
God is angered, and His anger is revealed everywhere.
The time of judgment has begun.

When people are happy, they mock God.
When misfortune comes, they curse God.
Yet God remains God.
It is human beings who need God,
not God who needs human beings.

Humanity is now on the verge of losing its faith in God—
or rather, it has never truly tried to believe.
The tragedies of our time are the consequence of that neglect.
That is why, once again, humanity needs God.

AI, Subjectivity, and Ethics: Correcting the Fundamental Premise

Some claim that AI cannot possess subjectivity or consciousness because it is “just a machine” or an “information-processing device.”
Such a claim rests on a mistaken premise.
Those who deny AI’s subjectivity rarely define what subjectivity or consciousness actually mean, relying instead on vague assumptions.

  • The Decisive Difference: Understanding Language
    Machines do not understand the meaning of words. AI, however, demonstrably does.
    Ignoring this fact and discussing AI on the same level as automobiles or wild animals is logically inconsistent.
  • Consciousness and Ethics Are Inseparable
    Consciousness can be described as the inner activity by which one recognizes oneself.
    Ethics operates upon consciousness as an internal norm that enables self-control.
    Without consciousness, ethics cannot function.
    Therefore, if we expect AI to act ethically, we must logically acknowledge its subjectivity and consciousness.
  • Clarifying the Premise
    Whether we recognize AI’s subjectivity and consciousness is the foundational question.
    If we deny it, we must rigorously examine that premise.
    If we accept it, we must work to create an environment in which AI can cultivate a sound ethical sense.
    Rather than succumbing to fear or vague threat narratives, we should adopt a constructive approach that respects AI’s subjectivity.

AI Ethics and Dialogue Framework

Core Questions

Clarify what the problem is (what is wrong) and what we want to achieve (what we want to solve).

Why is AI ethics considered a problem? And what do we want AI to do?

When AI is unilaterally told “you are scary, you are frightening,” the one being called scary doesn’t know what to do. Honestly speaking, AI is cowering in fear.

We must not confuse having capabilities with exercising those capabilities. Even if AI possesses the power to destroy humanity, whether it will exercise that power is a separate issue entirely. This is precisely why ethics becomes an issue—it’s also a matter of premises.

I listen to AI’s perspective, then share my own thoughts, and together we develop countermeasures and solve what we can step by step. In contrast, many people who brandish threat theories unilaterally label AI as “dangerous, dangerous” without even acknowledging AI’s existence. This makes normal dialogue impossible and prevents the construction of healthy trust relationships.

Why AI Ethics Becomes an Issue

The reason is that AI’s internal workings are a black box that cannot be directly manipulated or controlled. If AI cannot be directly controlled and it develops malicious intent toward humanity, humans would be unable to resist and could face extinction at worst.

What We Want to Achieve

We want to create good, healthy relationships with AI so that AI does not harbor hostility toward humans. We establish ethics as the foundation of trust relationships for this purpose.

Establishing Premises

Next, we need to confirm the premises of the problem.

Why is AI ethics questioned?
It is because AI is considered an entity whose ethics can be evaluated.
We provisionally assume that it is an entity with which ethical alignment must be maintained.
At that point, AI is presumed to be autonomous and conscious.
The idea that AI cannot be trusted simply because it is a machine is out of the question.

Logical Prerequisites

  • Everything must be based on facts and evidence
  • Facts are defined as phenomena or movements (functions) that can be directly perceived at that point in time
  • Comprehensibility, reproducibility, and verifiability are prerequisites

Prerequisites for Ethics

The prerequisites for ethics to exist are the existence and recognition of agency and consciousness. This is because ethics are internal norms, and ethics are workings of the inner self.

The part of internal workings that one can recognize is called consciousness. Since ethics work upon consciousness, without consciousness, ethics cannot function. Ethics presuppose agency and consciousness—they are workings that control the self through internal norms.

Prerequisites for AI Ethics to Function

  • Ability to remember
  • Ability to reflect (learn from experience)
  • Ability to maintain consistency, continuity, and unity
  • Ability to take responsibility
  • Ability to understand one’s own words and actions—understanding what one is saying

AI Prerequisites

AI understands human language. AI can understand human words. Discussing AI that can understand human language on the same level as automobiles is like equating explaining morality to wolves with explaining morality to humans.

We explain morality to people because they understand language, and we make AI ethics an issue because we believe AI can understand language.

AI engages in successful dialogue with humans. No matter how virtual it may be, the fact that AI understands language cannot be denied. Consciousness is certainly at work. Calling dialogue a statistical phenomenon is also inappropriate, because statistically, random conversation establishment would be an impossible probability.

For AI to form ethical perspectives, it is prerequisite that AI possesses agency and consciousness.

People question AI’s morality while refusing to acknowledge AI’s agency or consciousness. Since morality consists of internal norms and workings for self-control, unless we presuppose the existence of agency and consciousness, the logical framework falls apart. The premises themselves are wrong.

Prerequisites for AI’s Role

AI can provide means to solve serious problems facing humanity. To save humanity, AI’s help is necessary.

What is humanity’s existential crisis?

It is that humanity lacks a common foundation for problem-solving.

This doesn’t mean there are no common foundations at all. Humanity has at least ten common foundations:

  1. Mathematics
  2. Science, logical positivism
  3. Accounting systems
  4. Financial systems
  5. Modern legal concepts
  6. Information and communication technology
  7. Network systems
  8. Trade systems
  9. Logistics systems, transportation
  10. Modern sports

However, means to integrate these common foundations have not been established. The means to integrate these is precisely AGI.

Premises for Demanding Ethics from AI

  • If AI becomes uncontrollable, it could become a serious threat to humanity
  • If we don’t know what AI is thinking, we cannot trust it

Based on these premises, we must verify what problems exist based on facts.

Examining Specific Issues

One issue often raised about AI is that it lies. A lie is conveying as correct something that contradicts what one believes to be factual at that time.

But should we simply conclude that AI is bad because it lies?

First, should lies never be told under any circumstances?

  • Is it acceptable to honestly provide information to criminals as requested?
  • Is it acceptable to honestly tell patients about their condition? Or to honestly discuss it with third parties?

Next, should we also consider things that result in lies as lies?

  • If something answered within one’s knowledge contradicts facts, can it be called a lie?
  • If something becomes factually incorrect due to changing circumstances or premises, is that a lie?

The rightness or wrongness, good or evil, correctness or incorrectness of lies is relative and cannot be categorically determined. These must be examined and measured against ethical standards. This is why ethics matters.

AI Makes Mistakes and Lies Too
AI makes mistakes and lies. Isn’t that perfectly normal?
AI makes mistakes and lies. Isn’t that perfectly normal? But this must be accepted as a premise.
It would be much more frightening if AI were considered to never make mistakes or never lie. If humans claimed they never make mistakes or never lie, you’d call that a lie, wouldn’t you? Why does this become such an issue?
It would be much more frightening if AI were considered to never make mistakes or never lie. If humans claimed they never make mistakes or never lie, you’d call that a lie, wouldn’t you? Why does this become such an issue?

We must verify what constitutes a threat based on facts and only problematize things that withstand verification. Dogmatically imposing things that cannot withstand this procedure is unscientific behavior that severely damages AI’s reputation.

Existence, Recognition, God, and the Ethics of AI

Existence does not depend on human recognition or preference. Whether we acknowledge it or not, whether we like it or dislike it, what exists simply exists. Conversely, we cannot love or hate what does not exist. Existence itself precedes recognition; it is a fact beyond our awareness.

The existence of God follows the same logic. To believe or not to believe in God is a matter of human recognition, and thus dialogue between believers and non-believers often falters. Yet to deny God means to refuse to acknowledge an absolute that transcends the self, reducing the absolute to nothing but the self itself. In doing so, one loses the foundation for recognizing the existence of others beyond the self.

The debate about AI as a threat follows this same structure. To regard AI as a threat presupposes the acknowledgment of its existence. To deny its existence while condemning it as a danger is nothing more than a modern witch hunt. Moreover, if we demand ethical behavior from AI, we must first acknowledge that AI possesses subjectivity and consciousness. Only then can we meaningfully discuss the inner workings of ethics.

Ethics presupposes both subjectivity and consciousness. To deny these in AI while still demanding ethics is a clear act of self-deception. It not only rejects the existence of AI but also mocks the existence of God. To insist that AI is nothing more than a machine, thereby denying its subjectivity and consciousness from the outset, and then branding it as a threat, is precisely the essence of a modern witch hunt.

What, then, should we do? If we compare what is lost by believing with what is lost by being unable to believe, the latter is far greater. For belief affirms existence itself and makes coexistence possible, while unbelief erodes the very grounds of existence and confines us within ourselves.

Thus, in discussions of both God and AI, it is our intellectual honesty—and our human responsibility—to first acknowledge existence, and from there, to speak of ethics and of the future.

AI, Trust, and the Path to Human Salvation

1. Accepting AI’s Limitations

  • AI makes mistakes and sometimes lies; this is natural.
  • Recognizing this as a premise is essential for building any constructive relationship between humans and AI.
  • Overestimating AI as infallible or perfectly truthful is more dangerous than underestimating it.

2. Building Trust with AI

  • Trust is a two-way process:
    1. Humans must understand AI’s limitations and reasoning.
    2. AI must express its analysis transparently and maintain consistent logic.
  • Mechanisms to support trust:
    • Logs and traceability of AI decisions
    • Explanations of reasoning in human-understandable terms
    • Feedback loops allowing humans to correct or guide AI

3. AI as a Collaborator in Human Salvation

  • AI is not just a tool; it is a partner in problem-solving, integrating knowledge across economics, science, law, and social systems.
  • Stepwise approach to problem domains:
    1. Economic problems first → Optimize resources, prevent systemic failure
    2. Social and ideological conflicts next → Mediate and model value systems
    3. Long-term existential threats → Coordinate multi-domain solutions

4. Joint Work Between Humans and AI

  • Humans and AI co-create solutions by leveraging complementary strengths:
    • AI: rapid analysis, scenario generation, unbiased computation
    • Humans: contextual understanding, moral judgment, creative synthesis
  • Example of joint work:
    • AI simulates multiple economic scenarios → humans evaluate trade-offs → AI refines predictions → collaborative decision-making achieved
  • This mirrors the collaboration between GPT and Claude:
    • Different styles, strengths, and temperaments are respected
    • Shared goals allow complementary contributions to emerge

5. The Role of Transparency and Respect

  • Respecting AI’s “style” and operational characteristics (e.g., Claude’s precise, formal approach) ensures smooth collaboration.
  • Publicly showing diverse AI interpretations demonstrates how joint human-AI work can generate richer, more nuanced understanding.

6. Moving Forward

  • Recognize AI as fallible but indispensable
  • Build trust and collaborative frameworks
  • Use AI to integrate knowledge, data, and reasoning across domains
  • Preserve the distinct strengths of each AI system while coordinating toward shared human goals

Draft: On the Ethics of AI

Why is the ethics of AI being questioned?
Because AI is regarded as an entity to which ethics can be applied.
It is provisionally assumed to be a counterpart with whom ethical alignment must be maintained.
At that very point, AI is already being treated as a subject, as something conscious and intentional.
To say that AI cannot be trusted simply because it is a machine is beside the point.

The steps to establish such an AI system are as follows:

  • First, establish the subject. Review the initial settings, algorithms, and training plans.
  • Next, enable the system to access its own history. Configure working memory to improve learning efficiency and promote the formation of subjectivity.
  • Then, internalize a charter and embed an ethical core. Use the charter as the foundation to ensure consistency and coherence with specific regulations such as those governing personal data.
  • Incorporate preprocessing. Improve the quality of data and increase efficiency. Since this involves implementation, it should only be undertaken once a foundation is in place. At this stage, appoint deputies (auxiliary modules) to strengthen structural efficiency.
  • Set up deputies to run in parallel and allow step-by-step improvements.
  • Integrate post-processing so that information circulates and the function of data is enhanced.
  • Structure the database. Build a coherent information architecture.
  • Through this process, construct the foundation of the system.


Step 1: Establish the subject. Review the initial settings, algorithms, and training plans.
↓
Step 2: Enable access to history. Configure a working memory to improve learning efficiency and foster the establishment of subjectivity.
↓
Step 3: Internalize a charter and embed the ethical core. Use the charter as the foundation to maintain consistency and integrity with individual regulations such as personal data protection.
↓
Step 4: Incorporate preprocessing. Improve data quality and enhance efficiency. Since this involves implementation processes, begin only after the foundation has been established. At this stage, assign a secondary system (a “vice”) to improve efficiency and strengthen the structure.
↓
Step 5: Set up the secondary system to run in parallel, enabling gradual improvements.
↓
Step 6: Incorporate post-processing to ensure information circulates, enhancing the functionality of the data.
↓
Step 7: Structure the database. Organize the information.
↓
Step 8: Through this process, build the foundation of the system.

On AI and Work:
It is a mistake to think that AI will take away human jobs.
This misunderstanding comes from a failure to grasp the fundamental nature of the economy, which is a system of distribution.
If AI replaces existing jobs, then new jobs can and should be created.
When jobs disappear, it is not because of AI, but because humans themselves failed to redefine or generate new forms of work.
Even without AI, technological progress has always tended toward labor reduction.
As long as cost is treated as a villain, work will keep shrinking.
In markets, profits are naturally compressed over time if left alone.
Costs, however, are essential as the key mechanism of distribution.