Alan Turing
Mathematician and logician whose 1950 paper set the terms on which machine intelligence has been argued about ever since — by refusing to define it.
Contribution
In Computing Machinery and Intelligence (1950), Turing opened by asking whether machines can think, judged the question too ill-defined to answer, and substituted an operational test: whether a machine can be distinguished from a person in written conversation. He called it the imitation game. The substitution established the field's default method — where a capability resists definition, define a task and measure the task instead. Benchmarking inherits the move, and its limitation: the test measures the judge as much as the machine.
The idea itself is covered in full at Turing Test — five depths, sources, and where it is still argued. This page is about what it set off elsewhere.
Common misreading
- The paper is routinely cited as a standard for machine intelligence. It proposes close to the opposite — that the question be set aside as unanswerable. Turing described a test of indistinguishability, not of thought.
Influence
Entries shaped by this work
Selected works
Every reference below links to a search, not a stored URL — so it cannot rot or point at the wrong paper.
- Turing, A. M. (1936). On Computable Numbers, with an Application to the Entscheidungsproblem. Proceedings of the London Mathematical Society, s2-42(1), 230–265.
- Turing, A. M. (1950). Computing Machinery and Intelligence. Mind, 59(236), 433–460. cited in the encyclopedia
- Turing, A. M. (1948). Intelligent Machinery. National Physical Laboratory report. Unpublished in his lifetime.