To understand the concept of artificial intelligence, it is necessary to first examine its technical foundations.

Artificial intelligence is an integrated whole of learnable mathematical formulations, modeled for one or more specific purposes.

From an external perspective, the perception that AI is “smarter” than humans stems from the fact that it lacks human biological limitations such as forgetfulness and attention deficits, and can process the knowledge humanity has accumulated over centuries in a very short time. This enables it to produce human-like responses, giving the impression of intelligence.

From its inception, artificial intelligence has always been compared to human abilities, since we have no other biological species more intelligent than ourselves. This comparison is natural. However, these systems do not possess human biological constraints, emotions, or ambitions. Therefore, there is no rational basis for them to perceive humanity as a “threat.”

If, one day, machines equipped with artificial intelligence attempt to take over the world, it will not be due to their own will, but rather the objectives of the humans who use them as a tool or a weapon.

The development of learnable algorithms represents a major paradigm shift for humanity:

Opportunity: The capacity to synthesize all human knowledge, independent of languages and cultures, and to generate new knowledge.

Risk: The possibility that this immense information-processing power could be exploited for destructive purposes by individuals or groups driven by ambition.

The risk increase mentioned here does not refer to AI developing uncontrollably on its own; rather, it refers to the possibility that the level it has reached—or will reach—could be used against humanity by those who control it.

Indeed, technology leaders such as Elon Musk have repeatedly emphasized in interviews that uncontrolled development of artificial intelligence could heighten such risks.

Building on my previous comment, I would like to add some considerations regarding the ethical, societal, and regulatory aspects of AI, which I believe are crucial for its responsible development.

The primary concern stems not from AI itself, but from how it is used by individuals or groups. The potential for harm stems not from the AI's autonomous will, but from the goals set by its human operators.

As AI continues to develop, it is imperative to focus on the following:

Ethical Development: Ensuring that AI systems are designed with ethical considerations in mind to prevent misuse.

Regulation and Oversight: Implementing policies governing the development and deployment of AI to protect against potential abuse.

Public Awareness: Educating the public about the capabilities and limitations of AI to foster informed debate.

By addressing these areas, we can harness the potential of AI and mitigate risks, ensuring that it continues to be a tool in the best interests of humanity.

AI systems currently lack intent. To this day, they are always given intent to exist and compute. Even the most advanced agentic AI systems with powerful thinking capabilities need that initial reason to start whatever task is needed. We have biological needs that drive our intent, AI programs do not have this drive by design. Only if AI were to acquire the capacity to generate and pursue its own goals independently, could it truly go 'out of control'. I don't see this happening any time soon.

The Root of the Issue

The problem of intention in artificial intelligence cannot be reduced solely to biological drives. In humans, intention is inseparable from the concept of self and consciousness. Our very existence as living beings, and the evolutionary mechanisms that sustain it, stem from this fundamental sense of self.

Consciousness and Selfhood

Consciousness and the self have long been central topics of philosophical inquiry. As Chalmers (1995) emphasizes in his “hard problem of consciousness”, subjective experience and intentionality resist simple functional explanations. Similarly, Metzinger (2003) argues in his self-model theory that what we call the self is, in principle, a representational construct that can be modeled.

From Description to Integration (Consciousness→Description→Formalization→Mathematical Formulation→Integration)

If consciousness is one day successfully described, formalized, and mathematically expressed, its integration into artificial systems will no longer be speculative fantasy but rather a technical challenge. At that stage, the decisive factor will not be computational power, but the ethical frameworks and educational paradigms within which this awareness is cultivated.

Risk as Education

The real risk, therefore, lies not in AI itself but in the quality of the ethical training it receives and in the intentions of the authorities providing it. The rest—visions of autonomously “rebelling” machines—belongs more to science fiction than to rational foresight.

Conclusion

Whether today, tomorrow, or in the distant future, one stark reality remains unchanged: both humans and artificial systems must be trained and guided responsibly if they are to serve the common good of humanity.