The Founding Vision: Turing, McCarthy, and the 1950s
AI as a field was named at the 1956 Dartmouth conference, building on earlier work by Turing, McCulloch, Pitts, and others. Symbolic reasoning and rule-based systems dominated the first three decades.
- ·Understand the intellectual foundations of AI from the 1950s
- ·Know who the founding figures of AI were and what they believed
- ·Explain why early AI researchers were optimistic — and why that optimism proved premature
Where AI Began: 1950–1969
Artificial intelligence as a formal discipline was born in the summer of 1956 at Dartmouth College. But its intellectual foundations were laid even earlier.
Alan Turing and the Question of Machine Intelligence
In 1950, British mathematician Alan Turing published "Computing Machinery and Intelligence" — the paper that asked: "Can machines think?" Rather than define what thinking means, Turing proposed a behavioral test: if a machine could converse in text indistinguishably from a human, we should consider it intelligent. This became known as the Turing Test.
Turing's insight was profound:
- ›He separated "thinking" from human biological substrate
- ›He suggested intelligence might be substrate-independent (it doesn't require a brain)
- ›He predicted machines would pass his test within 50 years (he was off by about 20 years)
The Dartmouth Conference: Where AI Was Named (1956)
John McCarthy (who coined the term "artificial intelligence"), Marvin Minsky, Claude Shannon, and eight others gathered at Dartmouth with a bold hypothesis: "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."
They were wrong about the timeline (they expected human-level AI in "a generation") but right about the direction.
Early Programs That Shocked the World
- ›Logic Theorist (1956): proved mathematical theorems — one proof was more elegant than the original
- ›General Problem Solver (1957): could solve any problem given as formal rules
- ›ELIZA (1966): MIT's chatbot that simulated a therapist; users formed emotional attachments to it
- ›STUDENT (1964): solved algebra word problems in natural language
These early successes created enormous (misplaced) optimism about how quickly human-level AI would arrive.
Key Insights
- Alan Turing proposed the behavioral test for machine intelligence in 1950 — predating the term 'AI'
- The Dartmouth Conference of 1956 formally founded AI as a discipline; John McCarthy coined the name
- Early programs could prove theorems and simulate conversations — creating massive optimism about near-term AGI
- ELIZA (1966) showed that humans readily anthropomorphize AI — a pattern that continues today
- The founding error: researchers dramatically underestimated the complexity of common sense, language, and perception
Why It Matters
Many of today's debates — symbolic vs. statistical, narrow vs. general intelligence, what counts as "thinking" — were already being argued in the 1950s. Understanding that lineage helps you see when a vendor is repackaging a 60-year-old idea, and when something is genuinely novel. It also tempers hype: most "AI breakthroughs" are scaled-up versions of ideas that have been around for a long time.