Three Laws of Robotics
In plain wordsThree rules that science-fiction writer Isaac Asimov built into his fictional robots — do not harm people, obey orders, protect yourself — in stories that mostly show how such simple rules go wrong.
A fictional, strictly ranked set of constraints built into the positronic brains of Isaac Asimov's robots: first, a robot must not harm a human or, by doing nothing, let a human come to harm; second, it must obey human orders except where that would breach the first law; third, it must preserve itself unless that would breach either of the others. A 'Zeroth Law' placing humanity as a whole above any individual was added later. Asimov's plots turn on the rules' ambiguities — what counts as harm, conflicting orders, uncertain consequences — which has made them the classic teaching example of why short hand-written rules cannot capture human values. The same difficulties reappear in AI alignmentAI alignment, n.: Making an AI system genuinely try to do what the people it serves want — not a literal-minded, distorted or self-serving version of it. Also the research field working on this. as specification gamingspecification gaming, n.: When an AI meets the letter of its goal while defeating its purpose, like a genie who grants exactly the words of a wish instead of what the wisher meant., value specificationvalue alignment, n.: Building AI whose goals reflect human values — what people care about and consider right — rather than one that merely follows orders or chases a narrow target. and corrigibilitycorrigibility, n.: An AI's willingness to let people correct it — to accept being paused, retrained, overruled or shut down — rather than resisting or quietly working around them..
Any proposal that amounts to the Three Laws with extra steps will meet the same edge cases Asimov wrote about.
Origin
First stated in full in Isaac Asimov's short story 'Runaround' (Astounding Science Fiction, March 1942), after several robot stories in which such constraints were tacit; Asimov credited the editor John W. Campbell with helping formulate them. The Zeroth Law appeared in Robots and Empire (1985). Asimov's story 'Liar!' (1941) is also the earliest known use of the word 'robotics'.
Usage note
Often cited as if they were an engineering standard; no deployed AI system implements them. Natural-language rule sets such as the constitution in Constitutional AIConstitutional AI, n.: Anthropic's technique for making an AI less harmful by having it check and rewrite its own answers against a short written set of principles, so that people need not label harmful examples by hand. differ in being used as training signals rather than hard-coded constraints.
Related
- Isaac AsimovIsaac Asimov, prop. n.: A Russian-born American science-fiction writer and biochemist who invented the Three Laws of Robotics and is credited with the first use of the word 'robotics'.
- AI alignmentAI alignment, n.: Making an AI system genuinely try to do what the people it serves want — not a literal-minded, distorted or self-serving version of it. Also the research field working on this.
- corrigibilitycorrigibility, n.: An AI's willingness to let people correct it — to accept being paused, retrained, overruled or shut down — rather than resisting or quietly working around them.
- specification gamingspecification gaming, n.: When an AI meets the letter of its goal while defeating its purpose, like a genie who grants exactly the words of a wish instead of what the wisher meant.
- Constitutional AIConstitutional AI, n.: Anthropic's technique for making an AI less harmful by having it check and rewrite its own answers against a short written set of principles, so that people need not label harmful examples by hand.
- HAL 9000HAL 9000, prop. n.: The calm-voiced spaceship computer in the 1968 film 2001: A Space Odyssey that turns against its crew; still the best-known picture of a machine putting its goals ahead of human lives.
- Frankenstein complexFrankenstein complex, n.: Isaac Asimov's name for the fear that intelligent machines will turn on the people who made them, after Mary Shelley's novel about a scientist destroyed by his own creation.
Sources
- Asimov, I. (1942). Runaround. Astounding Science Fiction, March 1942.
Mentioned in
robotrobot, n.: A machine that senses its surroundings, decides what to do and physically acts, moving itself or handling objects, with some independence from a human operator. Factory arms, robot vacuums and self-driving cars are all robots., HAL 9000HAL 9000, prop. n.: The calm-voiced spaceship computer in the 1968 film 2001: A Space Odyssey that turns against its crew; still the best-known picture of a machine putting its goals ahead of human lives., Isaac AsimovIsaac Asimov, prop. n.: A Russian-born American science-fiction writer and biochemist who invented the Three Laws of Robotics and is credited with the first use of the word 'robotics'., Karel ČapekKarel Čapek, prop. n.: A Czech writer whose 1920 play about artificial workers who rise up against their makers gave the world the word 'robot'., machine ethicsmachine ethics, n.: The effort to build AI systems that can tell right from wrong and act ethically themselves, rather than only being used ethically by people., Frankenstein complexFrankenstein complex, n.: Isaac Asimov's name for the fear that intelligent machines will turn on the people who made them, after Mary Shelley's novel about a scientist destroyed by his own creation., human–robot interactionhuman–robot interaction, n.: The study of how people and robots communicate, cooperate and share spaces, and of how to design robots that people find safe, understandable, trustworthy and pleasant to live or work with.