Humanist AI Manifesto — Luxembourg Declaration on AI and Human Values
(source: Humanists International — Luxembourg Declaration on Artificial Intelligence and Human Values)
A humanist frame for AI governance — ten principles, stated as positive commitments:
- Human judgment — AI must never displace human judgment, human reason, human ethics, or human responsibility. Decisions that deeply affect people’s lives must always remain in human hands.
- Common good — AI should be a tool to serve humanity rather than enrich a privileged few; benefits should flow widely rather than concentrate power and wealth.
- Democratic governance — new technologies must be democratically accountable at all levels; no corporation, nation, or special interest should wield unaccountable power; regulators must build the expertise to keep pace.
- Transparency and autonomy — citizens can’t participate in democracies if decisions are opaque; transparency must be embedded in both laws and system design; everyone should control their data and be able to query/contest/reshape deployments.
- Protection from harm — foundational principle of all AI systems: prevent use in discrimination, manipulation, unjust surveillance, targeted violence, or suppression of lawful speech; long-term AI safety research must continue.
- Shared prosperity — technological advancement must not erode human dignity or entrench social divides; invest in training, education, and social protections for those most affected.
- Creators and artists — recognise and protect the unique value of human artists; IP frameworks must guarantee fair compensation, attribution, and protection.
- Reason, truth, and integrity — protect free inquiry, freedom of expression, and the health of democracy from misinformation, disinformation, and deliberate deception at scale.
- Future generations — decisions guided by responsibilities to future human generations, other animals, and the wider ecosystem; prioritise mitigation of environmental harms and long-term survival risks.
- Human freedom, human flourishing — the ultimate value of AI lies in its contribution to human happiness; embed shared values that promote human flourishing.
Why it’s in the vault
- It’s the “expanded Humanist AI manifesto” the owner asked to bring into mainstream discussion alongside the Asimov’s-laws thread from the Harari interview.
- It pairs with Asimov’s Three Laws (the robot-centric, law-shaped frame) as the human-centric, values-shaped counterpart — see Humanist AI Manifesto § Asimov below.
Asimov’s Laws — the comparison the owner flagged
“Asimov’s laws of robotics comes to mind… He was a smart guy — what did he think about humanism? In any case let’s bring that into mainstream discussion as well as the expanded Humanist AI manifesto.”
- Asimov’s Three Laws (from I, Robot, 1950): (1) a robot may not injure a human being or, through inaction, allow one to come to harm; (2) a robot must obey orders given by humans except where that conflicts with the First Law; (3) a robot must protect its own existence except where that conflicts with the First or Second Law. — Wikipedia: Three Laws of Robotics
- The difference that matters: Asimov’s laws are instructions embedded in the machine (a design constraint, and in his fiction repeatedly shown to be ambiguous, exploitable, and insufficient — the “Zeroth Law” had to be invented later). The Humanist Declaration is a contract on human society — a statement of the values and governance we owe each other, not a property of the robot.
- Asimov himself was a secular humanist (he was president of the American Humanist Association, 1985–92) — his stories consistently showed that laws alone can’t ensure good outcomes; human judgement and ethics must do the work. That is exactly the Humanist Declaration’s premise. This makes the two frames complementary, not competing: the Declaration is the answer to “what do we do when the Three Laws aren’t enough?”