1955: The Dartmouth AI proposal is dated

On 31 August 1955, John McCarthy, Marvin Minsky, Nathaniel Rochester and Claude Shannon finished and dated a short, ambitious document that would give artificial intelligence its name and its first organised research agenda. Titled “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence,” the seven-page text invited a handful of researchers to spend the summer of 1956 at Dartmouth College exploring the conjecture that “every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.”

The proposal was deliberately broad. It listed problems that still sound contemporary: automatic computers, how to program a computer to use language, neuron nets, theory of the size of a calculation, self-improvement, abstractions, and randomness and creativity. McCarthy coined the phrase “artificial intelligence” in the document itself, preferring it to earlier labels such as “automata studies” or “cybernetics.” The four authors—already established figures in mathematics, computing and information theory—signed the proposal and circulated it to the Rockefeller Foundation, which later provided modest funding.

Although the actual six-week workshop did not begin until the following July, the 31 August 1955 date marks the moment the field was formally christened and framed. The meeting that followed brought together many of the people who would dominate AI for decades: Allen Newell, Herbert Simon, Arthur Samuel, Oliver Selfridge and others. From those discussions emerged early programs such as the Logic Theorist and the General Problem Solver, as well as lasting debates about symbolic reasoning versus learning from data.

The intellectual thread that began with that dated proposal runs straight into today’s landscape. Large language models, reinforcement-learning agents and neuro-symbolic systems are all attempts to fulfil the 1955 conjecture that intelligence can be made precise enough for machines to simulate. When researchers now speak of “AI,” they are still using the name and the open-ended research programme first set down on a late-summer day in 1955.