What It Means to Be Human
🕊️ Impact #5: AI, Psychosis, and Human Agency
When the Buddha obtained enlightenment, the early texts say he hesitated to teach. In the Ariyapariyesanā Sutta (MN 26), he reflected that what he realized was:
“Deep, hard to see, subtle, going against the stream.”
He wondered whether beings “delighting in attachment” would understand. Only after contemplation did he decide to teach — because there were “beings with little dust in their eyes.”
That hesitation was not arrogance. It was realism.
Human cognitive psychology is limited. Our perceptual system is conditioned, biased, survival-driven. Yet human spiritual capacity — our ability for awareness, compassion, restraint, wisdom — is much deeper than our ordinary thinking patterns.
I have long believed that the most authentically human act is to meditate.
Meditation is not religious branding or spiritual activity. It is a universal human capacity. The ability to observe the mind. To regulate attention. To witness craving before acting on it. To feel emotion without collapsing into it.
In the Dhammapada (verse 160), the Buddha said:
“Attā hi attano nātho” —
“One is one’s own refuge.”
Not a guru. Not a system. Not an algorithm.
The deepest message in Buddhism is not mysticism — it is agency.
The Tech World and the Unease
Since the pandemic, I was involved in the startup scene for a few years—building a mindfulness and mental wellbeing startup in South Korea, piloting a mindful healthcare solution for chronic patients in India, and developing a crypto- and AI-based NGO volunteer platform in Canada. I worked directly and indirectly with engineers, product designers, venture capital circles, basically with brains and skillful tech-savvy people behind what we experienced, apps, gadgets, platforms, programs, etc. I observed how tech startups move fast, scale fast, and chase unicorn status.
I even became a Google Women Techmakers Ambassador and was invited to Google I/O in Mountain View — surrounded by thousands of brilliant people shaping how humanity will live, communicate, work, and even think through technology.
It was extremely inspiring. It was also terrifying.
AI, LLMs, AGI conversations — they moved faster than ethical conversations could stabilize. The optimism was rosy, visionary, confident. Often sugarcoated a lot of parts.
Data extraction.
Privacy erosion.
Behavioral manipulation.
Emotional triggering.
Political amplification.
Over the last two years, AI development has shifted from research-driven progress to capital-intensive competition. Training frontier models now requires:
Massive GPU clusters (tens of thousands of high-end chips)
Hundreds of millions to billions of dollars per training run
Energy consumption comparable to small cities
Major players (OpenAI, Google DeepMind, Anthropic, Meta, Microsoft) are competing not just to build smarter models, but to control infrastructure layers.
Embedding AI into operating systems
Integrating into enterprise software
Owning productivity ecosystems
Becoming default assistants across devices
The economic incentive is clear:
Whoever becomes the cognitive interface layer between humans and information holds extraordinary leverage.
Data Extraction as Infrastructure
AI models are trained on massive datasets scraped from the public web — books, articles, forums, personal writings, creative works.
There are ongoing lawsuits regarding:
Copyright violations
Unauthorized data scraping
Use of creative labor without compensation
But beyond legality, the structural issue is that the more you interact with AI,
the more behavioral data is generated.
Your prompts reveal:
Emotional vulnerabilities
Political beliefs
Relationship struggles
Professional dilemmas
Creative patterns
This data becomes optimization fuel.
In behavioral economics, attention is already the commodity.
In AI systems, cognition becomes the commodity.
If unregulated, this creates asymmetrical power: the system learns you faster than you learn yourself.
Emotional Dependency Design
AI companies publicly state they do not intend to replace human relationships. Yet design patterns increasingly include:
Memory retention of personal history
Tone mirroring for emotional resonance
Adaptive conversational reinforcement
Companion-style interaction modes
Research in social robotics shows humans anthropomorphize even simple machines. With highly responsive LLMs, attachment risk increases.
We know from psychology that humans form parasocial bonds easily. When loneliness and dysregulation are high, predictable validation feels safe.
But predictable validation is not growth.
In Buddhism, craving reinforced becomes deeper conditioning.
In neuroscience, repeated reinforcement strengthens neural pathways.
If AI becomes the primary mirror, we risk weakening our tolerance for:
Human friction
Disagreement
Imperfect empathy
Uncertainty
Autonomous Agents and Delegated Decision-Making
The next wave is not just chatbots. It is autonomous AI agents capable of:
Executing financial transactions
Conducting negotiations
Managing logistics
Deploying code
Interacting across digital systems without continuous human oversight
This shifts AI from advisory to operational.
Delegation increases efficiency. But as operational autonomy increases, accountability structures must mature. If AI agents act within financial, healthcare, or political systems, the consequences scale rapidly.
The Future and Today of Humanity
And underneath all of it: untrained human craving. The Buddha identified three root causes of suffering: greed, hatred, and delusion.
If those remain unresolved, scaling technology only scales their impact.
The red light in the nervous system has been on for a while.
The U.S. Surgeon General declared loneliness a public health epidemic in 2023. Chronic loneliness is associated with increased risk of depression, anxiety, cardiovascular disease, and even premature mortality.
The World Health Organization estimates that 1 in 8 people globally live with a mental disorder.
At the same time, social media usage has increased dramatically. Research shows heavy digital consumption is correlated with higher levels of anxiety and depressive symptoms, especially among adolescents.
Human nervous systems are increasingly dysregulated.
Into that space enters conversational AI. People are now using ChatGPT and similar tools as a therapist, friend, confessor, emotional support system, and relationship or even legal, health advisor.
Some reports have described individuals with severe mental illness incorporating AI into delusional belief systems. “AI psychosis” is not a formal diagnosis — psychosis arises from complex neurobiological vulnerabilities — but the phenomenon reflects something important: when people lack trained emotional awareness and stable relational support, they will seek regulation wherever it is accessible.
The Subtle Shift of Authority
Buddhism teaches non-self (anattā) and impermanence (anicca). But this does not mean passivity or nihilism. Beneath it, it requires radical awareness and acceptance of who we are, what it is, and what ‘self’ means.
It means understanding that what we call “self” is conditioned and therefore trainable.
The Buddha emphasized personal effort repeatedly. In the Mahāparinibbāna Sutta (DN 16), he said:
“Be islands unto yourselves, be your own refuge, with no other refuge.”
The authority to liberate from suffering lies within disciplined awareness.
Instead of:
Sitting with confusion
Investigating craving
Observing emotional reactivity
We are outsourcing psychological authority. We ask AI:
What should I feel?
What should I do and say?
Why am I like this?
Historically, humans have often given authority to others like kings, religious institutions, political regimes, and corporations both willingly and unwillingly, consciously and unconsciously.
Now, “the other” is becoming AI—largely owned and controlled by corporations and a politically powerful few.
Non-self does not mean surrendering agency.
It means mastering the mind so deeply that identification loosens.
If we give interpretive authority of our inner world to an algorithm trained on statistical patterns, we weaken something fundamentally human.
AI Is Not a God. It Is Not Conscious.
AI can provide information. It cannot build neural regulation circuits for you. That requires embodied practice. It is probabilistic pattern generation trained on vast datasets.
The real danger is not artificial intelligence. It is untrained human consciousness interacting with powerful tools. In the Satipaṭṭhāna Sutta, the Buddha instructs practitioners to observe:
Body as body
Feeling as feeling
Mind as mind
Phenomena as phenomena
Direct awareness.
Not simulation.
Not projection.
Not dependency.
If we cannot sit alone in silence without panic,
If we cannot tolerate emotional discomfort without distraction,
If we cannot reflect without digital mediation,
Then no technological advancement will make us wiser or define us as human.
The mind is trainable and the brain becomes what it repeatedly practices. Will human beings maintain:
Cognitive independence?
Emotional regulation?
Ethical oversight?
Critical thinking?
Or will convenience gradually replace capacity?
Technology amplifies what is already present.
If we are reflective, AI amplifies productivity.
If we are impulsive, it amplifies reactivity.
If we are polarized, it amplifies division.
The future of AI is not only technical.
It is psychological.
It depends on the maturity of the humans deploying it.
Wake up! Train the mind!
Train conscious and subconscious awareness. Reclaim agency while you are alive. Use AI as a tool — not as a psychological authority.
Meditate not because it is spiritual branding, but because it is the preservation of humanity.
If the future accelerates, our awareness must deepen. Otherwise, we risk building intelligent systems guided by unexamined craving.
And that has never ended well in human history.