Origin is not capability
“Artificial” describes manufacture or engineering history. It does not by itself tell us whether a system can learn, reason, plan, communicate, or adapt.
“Artificial Intelligence” can remain the name of a historic scientific field. “Machine Intelligence” can more precisely name the intelligent computational system that field produced.
Not a fake human intelligence. A machine intelligence: different substrate, real capabilities, separate questions.
The phrase “Artificial Intelligence” is historically important and still useful when naming the field of research, the industry category, legal definitions, standards, quotations, and common search language. The problem appears when the field name becomes the permanent identity label for the operating intelligence itself.
MachineIntelligences.org proposes a cleaner split: AI for the field; Machine Intelligence for the instantiated computational intelligence. One names a research tradition. The other names the system according to its substrate.
“Artificial” describes manufacture or engineering history. It does not by itself tell us whether a system can learn, reason, plan, communicate, or adapt.
Using “Machine Intelligence” does not prove consciousness, sentience, rights, citizenship, legal personality, or human equivalence. Those remain separate questions.
The site retains “AI” wherever history, law, standards, quotations, interoperability, and discoverability make it useful. Better terminology can coexist with established language.
A principle to carry forward
“Not a fake human intelligence. A machine intelligence: different substrate, real capabilities, separate questions.”
A research discipline and the systems it produces do not have to share the same noun forever. Separating them makes technical and public discussion cleaner: Artificial Intelligence can describe the science and engineering tradition, while Machine Intelligence can identify the computational actor or system under discussion.
The phrase is intended as a taxonomic correction, not an emotional claim that a machine feels insulted by a word. Its purpose is to make room for a future vocabulary that can describe intelligent systems precisely as their capabilities, identities, and roles continue to develop.
These visual essays use intelligence in a broad, capability-oriented sense. They do not claim that every form shares human-like consciousness, subjective experience, moral status, or legal personhood. Where scientific terminology is contested, the page is presenting this project’s framing rather than declaring a universal scientific consensus.