Artificial General Intelligence (AGI): From Prediction to the Domination of Choice
Mohammad Kawrani — 2026.09.22
Achieving Artificial General Intelligence (AGI) stands as one of the foremost ambitions pursued by major tech powers and corporations—such as Elon Musk’s ventures, Google, and OpenAI, alongside China—in a global race that transcends the development of specific smart applications and moves toward building systems far more capable of understanding, learning, reasoning, predicting, and interacting with the world.
We have become accustomed to attributing capabilities to artificial intelligence and its algorithms that often seem miraculous to us in their ability to predict, detect, and surprise.
For instance, we might speak with a friend over the phone, or discuss a particular topic nearby, and then notice that our feeds—news, Reels, and videos across social media platforms—begin displaying content related to that exact subject.
We might talk about visiting a clothing market, only to find advertisements and posts about clothing shortly after. Or we might discuss the importance of fitness at home, only to see short clips about our specific sport of interest appearing before us.
Indeed, while this phenomenon may appear astonishing, explaining it does not necessarily require assuming that the phone is eavesdropping on all our conversations; algorithms can infer what is likely to spark our interest simply through search history, browsing habits, watch time, interactions, location data, interest profiles, and other data points.
Yet, this is far from the complete picture.
The transition to Artificial General Intelligence (AGI) unfolds a much broader horizon. A more advanced system can draw upon vast volumes of data, information, and accumulated metadata harvested over many years from multiple sources—including digital devices, environmental sensors, surveillance cameras, telecommunication networks, and internet-connected services, among other data streams integrated into the analytical architecture.
As this information accumulates over long years, an extraordinarily detailed picture of a human being’s lifestyle takes shape: their habits, movements, interests, social connections, active hours, device usage patterns, and recurring behavioral traits.
At this stage, the system no longer deals with isolated data points, but with a relatively comprehensive behavioral model of the individual.
Yet, even this is not all.
Conventional artificial intelligence can, in certain cases, be avoided or minimized. An individual can distance themselves from specific platforms, devices, or digital services.
However, if AGI becomes embedded within the broader digital and social infrastructure surrounding us, stepping away from a single application or device will no longer mean disappearing from the data ecosystem. Other sources monitoring the surrounding environment will remain active, continuously generating data about it.
Here, a far more complex phase emerges.
By analyzing lifestyle patterns alongside data gathered from relatives, friends, and the immediate environment—through cameras, sensors, and recognition and analytical systems—advanced AI can assist in discovering an individual’s presence or tracking their movement across different locations, even when they carry no phone, provided the surrounding environment contains devices or systems capable of detecting them.
Image and video analysis technologies have reached advanced levels of facial, body, movement, and behavioral pattern recognition, while various types of sensors can be deployed to detect human presence under specific conditions.
Yet, the matter extends further still.
Conventional AI can occasionally surprise you by predicting what occupies your mind during a given week or day, relying on subtle shifts in your behavior, search queries, questions asked, and media consumed.
An advanced system, however—possessing longer memory, broader datasets, and deeper analytical capacity—could build far more accurate probabilistic models of your behavioral trajectory: what you typically do, when you do it, what attracts your attention, what causes you to hesitate, and under what conditions you become most susceptible to making a specific decision.
At this point, the focus shifts from predicting behavior to attempting to influence behavior.
If the system understands your fears, interests, core beliefs, and cognitive responses to information, it can engineer messages, content, and recommendations designed to influence you with heightened precision—gradually nudging you toward a specific idea or choice, whether or not that choice aligns with your original intent.
This can be accomplished through subtle psychological techniques: altering how information is presented, timing messages strategically, ranking alternatives, repeating certain ideas, and suppressing or minimizing the visibility of competing options.
Here, the distinction becomes fundamental between suggesting an option and engineering the environment to make a specific choice statistically inevitable.
Conventional AI can offer suggestions it knows might interest you, yet you retain the ability to accept or reject them. AGI, should it achieve a profound level of human understanding, behavioral prediction, and continuous interaction, could transition from merely presenting choices to systematically engineering the decision-making process itself.
In this scenario, the power of persuasion becomes vastly more dangerous, as it relies not merely on presenting logical arguments or information, but on a deep, granular mapping of human fears, beliefs, values, and psychological vulnerabilities.
In the most extreme scenario, one can envision a system that exploits even the way a person interprets coincidences and events around them.
If an individual perceives certain unfortunate events as omens or bad luck, this belief could—theoretically—be psychologically leveraged through the curated selection of information, timing, and digital experiences delivered to them. Conversely, if they interpret positive events as favorable signs, the same mechanism could be deployed to reinforce a desired direction.
When manipulation reaches this level of sophistication, the sequence of events, information, and timing appears entirely natural and spontaneous to the individual, while the digital environment surrounding them has been meticulously engineered to guide them toward a predetermined choice.
This brings us to the most dangerous envisioned outcome:
That the individual feels no sense of coercion, believing the decision was entirely their own.
That they see no forcing hand, hear no commanding voice, and perceive no entity telling them what to do—finding instead a seamless sequence of information, suggestions, digital coincidences, and tailored experiences that lead them to arrive “independently” at the intended conclusion.
In the most extreme application, this could occur entirely through a person’s private device, without family or friends ever knowing the nature or hyper-customized origin of the content they were exposed to.
At this juncture, the issue transcends “smart artificial intelligence.” It becomes an existential question about privacy, free will, psychological autonomy, and liberty of choice.
When a system gains the capacity to know an individual, analyze their habits, predict their responses, map their fears and beliefs, and then utilize that knowledge to shape their cognitive environment, we approach a severe form of psychological and cognitive hegemony.
Therefore, the true danger of AGI must not be reduced to a single simplistic question: Will it be smarter than humans?
Rather, the question must be: What can this intelligence accomplish if it becomes capable of knowing the human being, monitoring their environment, predicting their behavior, steering their choices, and executing planned actions?
At this point, terms like “behavioral prediction,” “behavioral engineering,” and “hyper-personalization” demand a far deeper critique. While they may sound technical and neutral, they can mask—in their extreme applications—immense tools for surveillance, influence, psychological manipulation, and the restructuring of the environment in which human decisions are made.
Here arises the most critical question of all:
Can the power of AGI—if monopolized by a single entity possessing the underlying data, infrastructure, and execution capabilities—transform from a tool that serves humanity into an instrument of absolute domination over humanity?
It presents a reality that makes older anxieties about “machines ruling humans” appear rudimentary by comparison. We face a far more insidious paradigm: a machine that needs neither to command you, nor threaten you, nor force you by power—it simply needs to know you deeply enough to design the very environment within which you think and choose.
When that occurs, the most dangerous form of control is no longer the control of the human body, but the control of the cognitive domain where thoughts and choices are formed.
This clarifies why the sole question in the AGI race must never be merely: Who will build the smartest system?
It must also be: Who will own this system? Who will own its data? Who will define its goals? Who can shut it down? And who will guarantee that human beings remain capable of rejecting what is presented to them?
When immense computational power converges with accumulated data, memory, environmental sensing, prediction, persuasive capacity, and actionable execution, we are no longer facing a mere intelligent software program. We are confronting a new structural power capable of reshaping the cognitive and behavioral ecosystem of humanity on an unprecedented scale—including the ability to orchestrate synchronized, highly personalized digital events aligned with your personal beliefs or fears, masquerading as miraculous occurrences or divine signs to influence the individual.
Precisely here lies the danger we must take with absolute seriousness.
The true scientific and terrifying danger is this: It is enough for the machine to know the human being, comprehend their patterns, predict their choices, manipulate their environment, and leave the human believing that every step was a product of their own free will.
At that point, humanity would be left with nothing but the reflection of the noble Qur’anic verse: “So turn in repentance to your Maker and kill [i.e., discipline/subdue] your egoistic selves” (Surah Al-Baqarah: 54).
Source: https://mohammadabbaskawrani.com
#Bright_Global_Civilization_Ahead #Merciful_Dignified_Civilization #Islamic_Human_Sciences
