Why the 1956 Dartmouth Conference Still Matters

In the summer of 1956, a small group of researchers gathered at Dartmouth College for a workshop that would give a new field its name. The event produced no breakthrough system and solved no famous problem. Seventy years later, it is still the moment we point to when we ask where artificial intelligence began.

A proposal with a name

The workshop grew out of a 1955 proposal written by John McCarthy together with Marvin Minsky, Nathaniel Rochester, and Claude Shannon. It is where the phrase artificial intelligence first appears, and the choice was deliberate: McCarthy wanted a label that set the effort apart from neighbouring fields like cybernetics and automata theory, and that named a goal rather than a technique.

The proposal’s central conjecture was audacious — that every feature of intelligence could, in principle, be described precisely enough for a machine to simulate it. That single sentence still frames the debate today.

What actually happened

The reality was more modest than the ambition. Researchers came and went over the weeks; there was no shared agenda and no dramatic result. What the meeting did produce was a community and a direction. It put language, reasoning, abstraction, and learning on the table as problems a machine might one day handle, and it gave a scattered set of researchers a common identity.

Why it still matters

Two things make Dartmouth worth remembering. First, it established a pattern that would repeat for decades: bold promises, real but partial progress, and a hard limit that forces the next idea. Second, it framed the question that the field is still arguing about — whether intelligence is best built from explicit rules or learned from data.

If you want the longer version of that story, the evolution of AI traces it decade by decade, and the philosophy section takes up the argument Dartmouth started.