sciencebriefs
9:52in productionCh. 1 · The Founding Event/ 9:52 · ceiling 15 min
Computing & AI

Dartmouth workshop

1956

The Dartmouth workshop didn’t build AI — it invented the idea that AI could be built.

The Dartmouth workshop founded AI as a field by naming it and framing its ambition — not by delivering results.

Computing & AIArtificial intelligenceComputingAbstractionArtificial intelligence across the network →
Chapters & takeaways4
  1. 1:18
    The Founding Event

    This was the founding event of AI as a field — not a conference, not a paper, but the formal birth of a discipline.

  2. 3:07
    Where the Name Came From

    The term 'artificial intelligence' first appeared in the formal proposal dated 2 September 1955.

  3. 5:04
    The Conjecture

    The entire effort rested on a single, untested conjecture: that all features of intelligence are precisely describable and simulable.

  4. 6:24
    The Four Aims

    It aimed to make machines use language, form abstractions, solve human-reserved problems, and improve themselves — none of which it achieved.

Worth your time?

Yes. Study the whole thing.

4.5/ 5
What works
  • established AI as a named field
  • introduced the term 'artificial intelligence'
  • defined the scope of the field by explicit conjecture
  • set enduring research aims
What does not
  • produce working systems
  • report empirical data
  • validate the conjecture
  • achieve any of its four aims
Study it if
  • historians of science
  • AI researchers tracing disciplinary origins
  • students of computing history
Skip it if
  • practitioners seeking technical methods
  • engineers evaluating current AI capabilities
  • policy makers assessing deployed systems
The written brief1 min read

What the work claims

That every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can simulate it.

How it was done

The workshop was a formally proposed, two-month, ten-person summer study at Dartmouth College in Hanover, New Hampshire, in 1956.

What holds up

It established artificial intelligence as a field through formal proposal and collective framing. It introduced the term ‘artificial intelligence’ and defined its scope by conjecture.

What does not

It did not produce working systems, algorithms, or empirical results. It made no measurements, ran no experiments, and reported no data.

Why it matters beyond the lab

It set the conceptual agenda for decades: language use, abstraction, human-scale problem-solving, and self-improvement remain active research targets — not because they were solved, but because they were named and claimed as possible.

Is it worth your time

Yes — it is the origin point of AI as a named, self-conscious field, not a later milestone or technical advance.

Same field · Computing & AI4 of 46
9:04
Turing testAlan TuringThe Turing test is a precise, behaviour-first redefinition of intelligence—not proof of thought, but a standard for functional equivalence in dialogue. It endures because it is simple, testable, and ruthlessly limited.
13:00
Transformer (deep learning)2017The 2017 paper Attention Is All You Need, from a team of researchers at Google, proposed the transformer architecture, which replaced the step-by-step recurrent processing used by earlier language models with a self-attention mechanism letting every part of an input be processed against every other part at once, making the whole computation far more parallelisable on modern hardware. That parallelism is a major reason transformers could be scaled up to the very large models trained since, including BERT in 2018, the GPT series beginning that same year, and the wave of systems from GPT-3 in 2020 through ChatGPT's public release in late 2022 and beyond. These large language models are typically built in two stages, an initial pretraining phase in which the model learns to predict the next word across enormous quantities of text, followed by a fine-tuning phase, often using human feedback, that shapes the model into a more instructable assistant. Even as capabilities have grown substantially, well-documented limitations persist, including confidently stated factual errors known as hallucinations, weaknesses in multi-step planning and reasoning, and reliance on human-selected training data whose biases the models can reproduce, and researchers remain divided over how much of what these models do reflects genuine understanding versus sophisticated pattern recombination.
13:00
Perceptron1957Frank Rosenblatt, working at Cornell Aeronautical Laboratory, built the perceptron, a machine that adjusted its own connection weights from labelled examples rather than being explicitly programmed. First simulated on an IBM computer in 1957 and then built as physical hardware, the Mark I Perceptron, by 1960, it drew wildly overstated press coverage about machines that would soon think and reproduce themselves. In 1969, Marvin Minsky and Seymour Papert published a proof that a single layer of the kind Rosenblatt had built could not learn certain patterns, most famously exclusive-or, and although their book acknowledged that multiple layers could solve the same problems, the narrower message was widely read as a verdict against neural networks generally, contributing to roughly a decade of reduced funding and interest before the approach was revived with multilayer networks trained by backpropagation.
9:19
Evolutionary computationEvolutionary computation is a family of global optimisation algorithms inspired by biological evolution. It originated conceptually with Alan Turing’s 1948 genetic search proposal. It emerged formally in the 1950s–1960s through independent developments: evolutionary programming (1962, USA), evolution strategies (1964, Germany), and genetic algorithms (1960s, USA). By the 1990s, the unifying term 'evolutionary computation' was coined, and genetic programming emerged as a fourth major branch. The approach is used both for engineering optimization and as an in silico tool in evolutionary biology.
Up next in Science

Higgs boson

2012 · 13:00

The 2012 discovery at CERN's Large Hadron Collider confirmed, at five-sigma significance, the particle that three groups of theorists had predicted in 1964, closing the last open piece of the Standard Model of particle physics.

Play nextBack to Computing & AI
13:00