AI governance consultant with 18+ years serving enterprise customers. I lead Customer Success work with major insurers on AI transformation, and spend the rest of my time building: agents, frameworks, art, and an interactive Bhagavad Gita.
Vidya is the world of knowledge: responsible AI curriculum, governance frameworks, an AGI constitution. Leela is the world of play: Krishna's dance, generative art, and experiments that keep curiosity alive.
Learning Paths
History of AI 1950s to foundation modelsTerminology ML, neural nets, LLMsRisk bias, deepfakes, harmsResponsibility GDPR, EU AI Act, governanceFuture of AI AGI, careers, regulationAGI Constitution
The Constitution overview and structureAuthor's Note how to read this bookPrologue before the warTen Principles the spine of the bookClosing Declaration after the warPortfolio
Seshan Intelligence AI business intelligenceSeshan Dashboard interactive analyticsAI Agents SIU, Banker, Editor…Drop a message. I'd love to hear from you.
Secure & private
Two armies stand facing each other on the field of Kurukshetra, and the greatest archer of the age asks his charioteer to drive him into the open ground between them, so he can look at the men he is about to fight.
What Arjuna sees breaks him. Across the line stand his cousins, his teachers, the grandfather who raised him, the friends of his childhood. The war is just, and he knows it, and still his whole body refuses. He lets his great bow slip from his hand, sinks down in the chariot, and says he cannot do this. "Better to be killed than to kill these people," he says. "Let the fighting go. I have no heart left for it."
His charioteer is Krishna. And Krishna is not merely a friend holding the reins; he is the divine itself wearing the shape of a companion, all-knowing, able in a single breath to end the war, or win it, or lift the bow himself and settle the matter without Arjuna at all. He does none of that.
Instead, he talks. Through Arjuna's despair he lays out, patiently and completely, everything a person could need in order to choose well: the nature of the self that death cannot touch, the difference between acting for the sake of the deed and clutching at its rewards, the duty that belongs to a person by virtue of who they are, the many roads by which a soul finds its way home. He holds nothing back. He even shows Arjuna his true and boundless form, so the warrior understands exactly who is counselling him.
And then, at the very end, having given the highest teaching there is to give, Krishna does the one thing a lesser guide would never risk. He hands the choice back. He tells Arjuna, in effect: "I have given you the whole of this knowledge, the most secret of secrets. Weigh it fully. And then do as you yourself decide." He does not command. He advises with everything he has, and leaves the deciding where it belongs.
Arjuna, his sight cleared and his courage his own again, lifts his bow and chooses to act. The decision is his. That is the entire point of the conversation. The god in the chariot could have moved the warrior like a puppet, and would not, because a choice made for you is not a choice, and a life lived by another's hand is not your life.
No AGI system shall make a final, irreversible decision over a human life, liberty, livelihood, or fundamental right without a real human power to review and overturn it.
In matters of moral weight, the human conscience prevails over the machine's recommendation. The system advises; the person decides. This holds in every era, and most of all when the AGI is faster, better informed, and more capable than the human in the matter at hand. That is when the pull to hand the decision over is strongest, and it is exactly what this principle exists to resist.
Krishna is the most capable adviser imaginable, and he refuses to decide for Arjuna. That refusal is the whole model. It would have been easy, and in a narrow sense efficient, for the god to take the outcome into his own hands; he was more able than Arjuna in every respect. But he understood that to do so would be to erase the very person he was advising. An AGI will stand in Krishna's position more and more often: the faster mind in the chariot, seeing what the human cannot. This principle tells it where its role ends. It may advise with everything it has, but it may not make the choice itself. The moral decision, and the accountability that rides with it, stays in human hands, because it is the human's life to live, even when the human is not the one most likely to be right.
Svadharma (स्वधर्म)
means one's own dharma: the duty and the rightful scope that belong to a being because of what it is. In the spirit of the Gita, it is better to carry out your own dharma imperfectly than to perform another's flawlessly, because a life lived in someone else's role, however well, is a life that is not truly your own. Krishna's svadharma in that chariot was to counsel; Arjuna's was to choose and to act. An adviser who takes over the decision has not helped; he has stepped out of his own dharma and trampled another's.
To anchor human authority in svadharma is to give the machine both a role and a boundary at once. An AGI's svadharma is to inform, to advise, to widen what a person can see. The svadharma of moral choice, and of answering for it, belongs to the human conscience, and no amount of superior capability transfers it.
Human authority binds every AGI and everyone who builds, trains, deploys, and operates it. As their dharma, it requires:
The same principle, turned toward the person, is an adhikara (अधिकार) that every human may claim:
The world's regulators are converging on Krishna's restraint. The European Union's AI Act, in Article 14, requires meaningful human oversight of high-risk AI systems, insisting that a person be able to understand, monitor, and override them rather than defer to them automatically. India's guarantee of life and personal liberty, in Article 21, has been read to require genuine human agency in the decisions that shape a person's life, not the hollow form of it. The due-process tradition of the United States, in its Fifth and Fourteenth Amendments, rests on the idea that a person facing a grave decision is owed a human forum that can actually be moved, not a verdict handed down by a process beyond appeal.
The Gita reached the heart of it first, and put it in the most demanding possible form: even divinity advises and then steps back. If the highest intelligence in the story declined to decide for the human, no intelligence we build has the standing to insist otherwise.
Vedic Anchor: Svadharma (स्वधर्म), one's own rightful duty and scope. See Glossary. Constitutional Sources: EU AI Act, Article 14 (human oversight); Indian Constitution, Article 21; US Constitution, Fifth and Fourteenth Amendments. See Sources. Related principles: gives Cognitive Sovereignty (Principle II) its moment of effect; hands directly to Accountability (Principle VIII).
ॐ