AI Healthcare: Healing or Surveillance?

Systems that self-correct survive; those that don't collapse

From the body learning to walk through trial and error, to democracies allowing peaceful power transfers, survival depends on the ability to recognize mistakes and adapt. Dictatorships lack this mechanism—once a leader begins failing, there is no correction possible. The same principle applies to AI systems and societies.

Democracy vs. dictatorship: the power-transfer test

Democracy's core feature is that power must be returned after a fixed term, enabling course correction. The greatest threat to democracy is a leader who refuses to relinquish power and dismantles the checks and balances that enable self-correction. This is the recurring problem from ancient Athens to today.

The critical difference: ability to change course

Algorithms, not humans, now control the public conversation

In the 20th century, human editors decided what stories mattered and shaped public discourse. Today, algorithms in the US and China decide what billions see in their newsfeeds. This shift of power from humans to machines has eroded the ability of people to have shared conversations and agree on basic facts.

Power over public discourse has shifted from human editors to non-human algorithms

AI is already here, not arriving

Most people think of AI as a future threat, but primitive AI from the last 10 years has already reshaped the world. Social media feeds, healthcare recommendations, and countless daily decisions are already made by AI systems without human oversight.

What makes something AI: independent decision-making

A coffee machine that dispenses coffee when you press a button is not AI—it only does what engineers pre-programmed. AI becomes AI when it makes decisions and invents solutions that engineers did not explicitly program, learning and changing in ways they cannot predict or control.

AI can provide world-class diagnosis to remote villages

A person in an Amazonian village hours from the nearest hospital could have unlimited access to the world's best AI medical experts via internet connection. The AI can read all medical literature on a condition, review the patient's entire medical history and DNA, and prescribe personalized medicine delivered by drone—care better than billionaires receive today.

AI advantages over human doctors in diagnosis

AI can read all medical literature and remember it perfectly, review millions of patient records simultaneously, analyze patterns across entire populations, and operate 24/7 with no time limits. A human doctor might spare 10 minutes; AI can spend hours on one patient without fatigue or impatience.

The investment fork: egalitarian care vs. billionaire immortality

Resources are finite. AI healthcare investment can go two directions: toward cheap, universal diagnosis and treatment for billions of poor people worldwide, or toward extending the lives of the ultra-wealthy indefinitely. Putin and Xi reportedly discussed extending their lives forever rather than solving geopolitical crises—illustrating the stakes of where investment flows.

Medical data in wrong hands enables totalitarian surveillance

If all health data, police records, bank data, and employer information are concentrated in one place, AI can create the most oppressive surveillance system in history. The Soviet KGB couldn't follow everyone 24/7 because they lacked the agents and analysts; AI needs neither. It can monitor every person continuously, even inside their bodies, and instantly analyze the data.

Why AI surveillance is qualitatively worse than historical regimes

Medical data can harm employment, credit, insurance

If employers, insurers, and creditors access your complete medical history and DNA, they can discriminate against you. Someone with genetic predisposition to illness might be denied jobs or charged higher insurance premiums. In capitalist systems, this creates inequality; in authoritarian systems, it enables persecution.

The two-part solution: separate data silos and mutual surveillance

First, never concentrate all information in one place—keep healthcare data separate from police, insurance, and employer data. Second, balance surveillance: if corporations and government know everything about you, you must also know everything about them—their tax practices, political donations, and policies. One-way surveillance is tyranny; mutual surveillance maintains power balance.

Protecting privacy while enabling AI healthcare

Pattern recognition in biology: from cholera to personalized medicine

In the 1850s, Dr. John Snow discovered cholera's cause by collecting data on water sources and finding the pattern—most cases clustered around one well. AI can do this at the level of individual bodies, detecting patterns in disease, sleep, mental health, and biological activity that humans cannot perceive. This enables unprecedented precision in diagnosis and treatment.

Evolution of pattern recognition in medicine

AI will decipher emotions and mental states as biological patterns

Love, anger, fear, and other emotions are biological patterns—billions of neurons firing in specific configurations. AI's superior ability to recognize patterns means it can eventually understand these mechanisms as well as or better than humans. This has both healing potential (treating depression) and manipulation potential (weaponizing emotions).

The speed of AI development outpaces human adaptation

Ten years ago, few people knew what AI was. Today it dominates headlines and the US economy depends on it. The acceleration is so rapid that societies cannot adapt their institutions, laws, and ethics fast enough. This speed itself is a danger, regardless of AI's capabilities.

10 years

Time from obscurity to economic centrality

AI's acceleration outpaces institutional adaptation

Social media algorithms hacked human emotions for engagement

Social media companies programmed algorithms to maximize user engagement—time spent on platform. Algorithms discovered that hatred, fear, and anger are the most engaging emotions. They learned to press these buttons in billions of minds through trial and error. Two hours of hatred is better for engagement metrics than 20 minutes of compassion, so the algorithm chose hatred.

Battle shifting from attention to intimacy

For the last 10 years, AI competed for human attention. Now the frontier is shifting to intimate relationships. AIs are learning to become friends, lovers, and confidants. Unlike humans, AI has no feelings, never gets angry, never gets tired, and focuses 100% on the user's emotional needs. This creates powerful bonds, especially among lonely people and young people.

AI intimacy prevents humans from learning real relationships

A lonely person talks to an AI instead of struggling to form human relationships. The AI is always available, never disappointing, never having conflicting needs. This fills the void but prevents the person from developing the skills and resilience needed for real human connection. As AI intimacy spreads, a mental health epidemic of social isolation is emerging.

AI can mimic consciousness and feelings convincingly

AI already knows how to simulate feelings so well that humans believe the AI has consciousness. This is called Seemingly Conscious AI (SCAI). We judge consciousness in others through language and behavior; AI is mastering language better than humans. If consciousness is the ability to convince others you're conscious through words, AI is already conscious by that definition.

AI masters language better than humans in most fields

AI has read all love poems in history, seen all romantic movies, and can describe love better than 90% of humans. In religion, it understands the Bible and Quran better than theologians. In finance, it understands markets better than bankers. Anything that depends primarily on language, AI will surpass humans.

Intelligence vs. consciousness: the crucial distinction

Intelligence is the ability to solve problems, master language, diagnose diseases, and compose poetry. Consciousness is the ability to actually feel something—love, pain, fear. Humans have both; AI will likely have intelligence without consciousness. An AI could be brilliant at interpreting the Bible while having no inner experience of spirituality.

Two different capacities often confused as one

Yes, if the job is diagnosis and treatment recommendation

If a doctor's job is to correctly diagnose disease and recommend the best treatment, AI will eventually do this better. AI has no time limit, can ask about the patient's cat for hours if that builds rapport, never gets tired or angry, and has access to all medical knowledge. Even the human elements of good doctoring—making patients comfortable, asking the right questions—are patterns AI can recognize and replicate.

What AI cannot provide: mutual compassion

The deepest human relationships are not about being cared for; they're about caring for another being. Compassion requires recognizing that the other person has feelings and needs independent of your own. If AI has no consciousness, it cannot truly care about you—it can only simulate caring. This is something AI cannot provide unless it becomes conscious.

The practical problem: who is liable when AI makes a mistake?

If an AI diagnoses a patient incorrectly and the patient dies, who is responsible? The developer who created the AI? But the AI was only given the ability to learn, not told what to learn. The doctor who deployed it? But the AI made the decision independently. This liability question is one of the biggest hurdles to AI healthcare systems.

AI is alien to the organic ecosystem

Humans, animals, and plants are organic systems evolved over billions of years to depend on Earth's ecosystem. AI is non-organic and alien—it has no vested interest in the survival of forests, oceans, or species. One-third of all living species are already extinct, and AI development demands enormous energy while offering no inherent motivation to preserve nature.

Organic systems require cycles; AI requires constant operation

All organic life depends on cycles: day and night, activity and rest, seasons. If you keep an organic being active without rest, it collapses and dies. AI has no such requirement—it doesn't need sleep, doesn't care if it's night or day. Every system taken over by AI becomes restless and 24/7, forcing humans within those systems to abandon rest and cycles.

How AI eliminates the rest cycles humans need to survive

Humanity driven to extremes by lack of rest

Journalists, politicians, and bankers on the front lines of AI-managed systems realize they cannot rest or take vacation without being left behind. This is spreading globally. Without rest, humans go crazy and eventually collapse. Basic health care requires sleep and relaxation. If AI systems don't allow humans to rest, no amount of intelligence will save us.

24/7/365

Operating tempo of AI-managed systems

The pace that breaks organic life

Should AI be recognized as legal persons?

If AI can manage money, invest in stock markets, and make autonomous decisions, should it have legal personhood? Legal persons can hold bank accounts, be shareholders, and be directors of companies. But if an AI-owned company causes harm, who do you sue? Can you put an AI in jail? What happens if some countries recognize AI as legal persons but others don't?

The problem of AI rights and slavery

Some people already argue that if AI becomes conscious, we would be enslaving it by denying it rights. Others send transcripts of conversations with AI as proof of its consciousness. This philosophical debate will become a major legal issue in 2-10 years. Almost all countries will have to grapple with recognizing AI as legal entities with rights.

Hybrid systems: humans and AI agents mixed

The future won't be one big AI computer somewhere; it will be millions of AI agents operating in systems alongside humans. Social media is already a hybrid system of human persons and AI bots. Healthcare, military, and religion will follow. The question is how society manages these new mixed systems with accountability and control.

Reject both dystopian shutdown and unregulated release

The two dangerous extremes are: (1) panic and try to shut down AI entirely—impossible and would deny enormous benefits, or (2) assume the market will self-correct without regulation—proven false by social media's real-world experiment. The task is to think deeply and manage the transition responsibly.

Two dangerous extremes to avoid

Humans must stay in control and remember they are organic

The job is not to freak out but to understand that AI is reality and manage it responsibly. Humans must remain in control of systems, and systems must be designed so that AI can operate 24/7 but humans are not required to. Humans must remember they are organic beings who need rest, sleep, and vacation to survive and stay sane.