Step by Step
D
Dartmouth Conference (1956) — the field is born
John McCarthy coins the term "Artificial Intelligence" at a summer workshop at Dartmouth College. This is the official starting point of AI as a named academic field.
Example: if an exam asks who coined the term "Artificial Intelligence" and in what year, the answer is John McCarthy, 1956.
A
AI Winters (1970s, 1980s) — funding collapses twice
Early AI research overpromised and underdelivered. When results failed to match the hype, government and industry funding dried up — this happened not once but twice, in two separate decades.
Example: expert systems in the 1980s generated a second wave of hype and a second winter when they proved too brittle and expensive to maintain.
D
Deep Learning Breakthrough (2012) — the modern era begins
AlexNet wins the ImageNet competition by a massive margin using deep convolutional neural networks trained on GPUs, proving deep learning could outperform older approaches at scale.
Example: before 2012, most image recognition used hand-engineered features; AlexNet's win convinced the field that letting the network learn its own features from data was superior.
A
LLM / AGI-era buzz (2017–present) — Transformers to ChatGPT
The 2017 Transformer architecture leads to BERT, then GPT, then ChatGPT, triggering an explosion of generative AI adoption and renewed (sometimes overhyped) AGI speculation.
Example: the jump from GPT-3 to ChatGPT (2022) is what brought generative AI into mainstream, everyday use for the first time.
Applied Walkthrough
1
An exam question asks: "Which event marks the beginning of the modern deep learning era?"
2
You recall the DADA sequence: Dartmouth (1956) is the birth of the field itself, not deep learning specifically.
3
The AI Winters are periods of decline, not breakthroughs — so they're not the answer either.
4
The correct answer is 2012 — the AlexNet/ImageNet breakthrough — which is specifically what kicked off the deep learning era that led directly to today's LLMs.
Exam Application
Exams frequently ask you to place a name or event into the correct era: McCarthy and the Dartmouth Conference belong to 1956 (the founding), Hinton/LeCun/Bengio (winners of the 2018 Turing Award) belong to the deep learning breakthrough era, and Transformers/BERT/GPT belong to the 2017-onward LLM era. Keep these three eras and their key names cleanly separated.
⚠ Common Trap
The most common trap is assuming AI is a recent invention because ChatGPT feels so new. In reality the field is nearly 70 years old, and the LLM boom is only the most recent of several major waves — the field has already survived two "winters" where progress stalled for a decade at a time. Don't assume today's excitement means there's no risk of another slowdown.
✓ Quick Self-Check
1. Who coined the term "Artificial Intelligence," and in what year?
John McCarthy, at the Dartmouth Conference in 1956.
Tap to reveal / hide
2. How many major AI Winters (funding collapses) have occurred historically?
Two — one in the 1970s and one in the 1980s.
Tap to reveal / hide
3. What 2012 event is considered the start of the modern deep learning era?
AlexNet's landslide win at the ImageNet competition, using deep CNNs trained on GPUs.
Tap to reveal / hide
4. What architecture introduced in 2017 eventually led to BERT, GPT, and ChatGPT?
The Transformer architecture.
Tap to reveal / hide
5. Which three researchers won the 2018 Turing Award for deep learning pioneering work?
Geoffrey Hinton, Yann LeCun, and Yoshua Bengio.
Tap to reveal / hide