AI & Tech One-page note
A Short History of AI
AI isn't new. It started with one question in 1950, went through two "AI winters", and only took off when computers learned from data. Here's the whole story on one page. Save this page.
Good to know
Simplified timeline. Sources: original papers, IBM, Open University. Checked Nov 2026.
What's on the page
The note as text, so you can read it, copy it or use a screen reader.
The start (1950s)
- 1950: Turing asks "Can machines think?"
- 1956: Dartmouth summer project on "artificial intelligence"
- Goal: machines that use language and learn
Early hopes, first "AI winter"
- 1966: ELIZA, an early chatbot
- 1969: a book on neural nets hits funding
- Early 1970s: big budget cuts (first winter)
Rules boom, second winter
- 1980s: hand-written rule systems
- Useful, but only in narrow areas
- Costly US and Japan projects fall short
- Funding and interest drop again
Learning from data
- 1997: IBM's Deep Blue beats chess champion Kasparov
- 2012: deep learning wins an image contest (15% vs 26% error)
- 2017: the "Transformer" design, used by today's chatbots
The chatbot era
- 30 Nov 2022: ChatGPT launches
- Now: one model, many tasks
- Still not human-level at everything
Sources
- Turing, A. M.: Computing Machinery and Intelligence. Mind, 1950
- McCarthy, J. et al.: A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence (1955)
- Weizenbaum, J.: ELIZA. Communications of the ACM, 1966
- Open University (OpenLearn): From boom to bust: the AI winter
- Chilton Computing (UKRI STFC): Lighthill Report, Artificial Intelligence: A General Survey
- IBM: Deep Blue
- Krizhevsky, A., Sutskever, I. & Hinton, G.: ImageNet Classification with Deep Convolutional Neural Networks (NeurIPS 2012)
- Vaswani, A. et al.: Attention Is All You Need (2017)
- IBM: What are large language models?
- Centre for Computing History: OpenAI launches ChatGPT
- International AI Safety Report 2026
How we checked this
Every fact on this note comes from the 11 sources listed here, checked against official or peer-reviewed sources where possible. AI tools help with drafting; a person checks every fact against the sources. How we research
Spotted a mistake? Tell us and we'll fix it and log it on the corrections page.