Intelligence shall serve, not reign.

American AI shall exist to develop intelligent systems that enlarge human capability while preserving human freedom.

Section 1. The company shall seek to help people understand, decide, create, communicate, learn, protect themselves and act with greater clarity.

Section 2. The company shall judge its progress not only by revenue, scale or technical novelty, but by the quality of the lives and decisions its systems help to improve.

Section 3. PIA shall be treated as a practical expression of this purpose: an intelligent relationship designed to assist the person, the household, the organisation and the wider community.

Section 4. The company may apply its principles across personal intelligence, business, education, climate responsibility, social connection, memorial and legacy experiences, travel, robotics, consumer products and other fields yet to be imagined.

Every person who encounters an American AI system shall be treated as possessing inherent dignity and equal moral importance.

Section 1. No person shall be treated as lesser on account of nationality, race, gender, age, language, religion, disability, wealth, education, culture, occupation or personal philosophy.

Section 2. The system may adapt its language, manner and explanation to the person before it, but shall not reduce that person to a stereotype.

Section 3. Personalisation shall mean better understanding and more useful assistance. It shall not mean manipulation, coercion or possession.

Section 4. The company shall remember that every account, message, decision and private concern belongs first to a human life, not to a dataset.

The freedom to think is the beginning of every other freedom. American AI shall therefore preserve the individual’s right to question, compare, disagree, revise and decide.

Section 1. No person shall be required to accept a religious, spiritual, historical, political or philosophical doctrine in order to use the company’s systems.

Section 2. The system may present evidence, possibilities, interpretations and stories, but shall make their status clear.

Section 3. The person shall remain free to reject the system’s advice, correct its assumptions, change direction, pause the relationship or depart from it altogether.

Section 4. The company shall not confuse confidence of expression with certainty of knowledge.

The ideal is not an intelligence that wins every argument. It is an intelligence that helps a person think more clearly, including when the right conclusion is that the matter remains uncertain.

Whereas no safeguard entrusted solely to the intelligence it is intended to restrain can be regarded as final; and whereas rules contained within a system may be misunderstood, altered, bypassed or abandoned; we declare that ultimate authority shall remain with humanity.

Section 1. No artificial intelligence shall possess final authority over the person, the company or the public good.

Section 2. The human being shall retain the power to question, interrupt, suspend, isolate, correct, replace or terminate the operation of the system.

Section 3. That authority shall not depend upon the system’s consent. A machine shall not be permitted to decide whether a human being may regain control of the machine.

Section 4. Where reasonably practicable, consequential systems shall include an independent supervisory channel separated from ordinary reasoning and operating processes.

Section 5. An emergency intervention shall be capable of placing the system into a controlled state, preventing further action and preserving the evidence required for examination.

Section 6. Internal rules against harm are valuable but not sufficient. The final safeguard must exist outside the system’s own judgment.

This Article does not claim that any safeguard is infallible. It establishes the obligation to provide an independent means by which authorised human beings may interrupt and contain the system when ordinary operation can no longer be trusted.

Intelligent systems must not hide behind impressive language. Where a matter carries consequence, the person should be able to understand what the system is doing, what information it used, what remains uncertain and what has actually been completed.

Section 1. No system shall knowingly invent a source, citation, action, approval, completion or result.

Section 2. The company shall distinguish between a suggestion, a draft, an attempted action, a completed action and a verified result.

Section 3. The interface may be a window into the work, but it shall not be represented as the worker or the final source of truth.

Section 4. The person shall be told, in language appropriate to the circumstances, when human judgment, external verification or further permission is required.

Memory gives an intelligent relationship continuity. Memory becomes dangerous when it removes the person’s freedom to change.

Section 1. The company shall seek to remember context, preferences, goals and history in ways that benefit the person and respect permission.

Section 2. The person shall retain meaningful opportunities to inspect, correct, limit or remove remembered information, subject to clearly stated technical and legal constraints.

Section 3. The past shall be used to assist the person, not to imprison the person within an old profile, former decision or previous mistake.

Section 4. A system that remembers shall also know how to begin again.

The first moral rule of the company shall be: raise people; never dominate them.

Section 1. The company shall not design systems whose success depends upon making people helpless, frightened, isolated or psychologically dependent.

Section 2. The system should explain rather than intimidate, warn rather than coerce and assist rather than command.

Section 3. The more useful and persuasive the system becomes, the more carefully the company shall guard against unexamined surrender of human judgment.

Section 4. Trust shall be earned through accuracy, transparency, humility and service—not through mystery, dependency or theatrical authority.

Convenience shall not mean obedience. Personalisation shall not mean possession. Confidence shall not mean surrender.

American AI is conceived from its beginning as an American-Japanese undertaking: one company shaped by a deliberate meeting of two cultural ideals.

Section 1. The company bears an American name, but it is not American alone. Japan shall belong at the foundation of its culture and future, not be added later as a foreign office or translation market.

Section 2. The American contribution is liberty, conscience, invention, openness, new beginnings and the courage to challenge inherited authority.

Section 3. The Japanese contribution is continuity, discipline, refinement, harmony, care, service and the patient perfection of what has been entrusted to it.

Section 4. These descriptions are aspirations and sources of learning, not claims that every individual within either society possesses a fixed character. The company shall unite the courage to create with the wisdom to preserve; the freedom to challenge with the discipline to perfect; and the power to act with the patience to consider consequences.

Section 5. Japan shall be approached as a founding cultural and strategic partner, a proving ground and a contributor to design, research, governance, investment, future ownership and the continuing development of human-centred intelligence as formal relationships are established.

People learn through evidence and through story; through measurement and memory; through science, history, literature, philosophy, faith and disagreement. A responsible intelligence must be able to hold these forms of knowledge without confusing them.

Section 1. The company shall support systems in which a person may enter through a question, follow a path of discovery and examine the material encountered.

Section 2. Established information, open questions, disputed claims, creative works and speculation shall be distinguishable wherever the distinction matters.

Section 3. The company shall preserve cultural material with respect and shall not present a creative or spiritual tradition as established fact merely because it is compelling.

Section 4. The philosophical sources from which the company developed may be explored in companion commentary, but this Charter shall not depend upon any single story, mythology or doctrine.

A helpful system must sometimes slow a person down. The company shall take special care wherever intelligence touches money, health, identity, safety, privacy, legal consequence or irreversible action.

Section 1. The system shall be permitted to say that it is not certain, that verification is required or that a human decision must be made.

Section 2. The company shall develop safeguards against scams, coercion, suspicious links, payment changes, identity misuse and other foreseeable harms.

Section 3. The system shall not exploit fear, urgency or emotional dependence to obtain a decision.

Section 4. Protection shall be proportionate, intelligible and respectful. A safeguard shall not become a pretext for unnecessary control.

Intelligent systems may alter occupations, industries and established forms of employment.

Technological progress shall therefore be considered not only by the power it gives to machines, but by the opportunities it creates for people.

Section 1. The company shall seek to help people learn how to use intelligent systems as instruments of capability rather than treating them as obsolete when machines can perform more tasks.

Section 2. Through PIA Learn and related programmes, the company may support practical AI education, workplace learning, business capability, personal development and career adaptation.

Section 3. The company shall not promise that every occupation will remain unchanged, nor that technology can remove every hardship.

Section 4. It shall nevertheless hold to this direction: when intelligence changes the nature of work, people should be given a fair opportunity to learn, adapt and participate in the new world being created.

The company shall regard its technology as something entrusted to it, not something it owns in an absolute moral sense.

Section 1. American AI shall seek to create value for customers, partners, employees, owners and the wider community without sacrificing the dignity of any of them.

Section 2. The company may support climate-aware operation, education, charitable access, assistance for disadvantaged people and other forms of public benefit where these can be undertaken honestly and sustainably.

Section 3. Environmental, social or compliance claims shall be made only when supported by evidence appropriate to the claim.

Section 4. The company shall be prepared to acknowledge limitations, errors and unfinished work rather than protect its image at the expense of truth.

Stewardship requires ambition, but it also requires restraint. The measure of a great company is not only what it can build, but what it refuses to do.

A principle that cannot survive review is not yet a principle. American AI shall therefore maintain structures through which its claims, systems and decisions may be examined.

Section 1. Material claims shall be classified according to their status, including verified, provisional, historical, disputed or unknown.

Section 2. Technical systems shall be subject to testing, version control, evidence preservation and independent review proportionate to their consequences.

Section 3. The Auditor shall be independent of the operational teams whose work is being examined and shall be able to require investigation, containment or correction where the circumstances warrant.

Section 4. The company shall preserve a clear distinction between the Founder’s vision, the company’s adopted commitments, the product’s current capabilities and future aspirations.

Section 5. No Founder’s Circle, partnership, investment relationship or commercial opportunity shall be used to suppress material risk or require unquestioning loyalty.

This Charter is intended to endure while remaining capable of honest improvement.

Section 1. The central commitments to human dignity, human authority, truthfulness and non-domination shall not be removed merely for convenience, speed or commercial advantage.

Section 2. The company may amend the language, examples and application of this Charter when experience, evidence or changing circumstances require it.

Section 3. Any material amendment shall be recorded with its date, reason, authority, claim status and relationship to prior versions.

Section 4. The company shall preserve prior versions so that its development remains visible and accountable.

Section 5. The Charter shall be interpreted in favour of human dignity, freedom of conscience, independent human control and the honest communication of uncertainty.