The Complete History of AI

From Turing Test to GPT: Explore the Full Journey of Artificial Intelligence

Explore the technology tree Browse the timeline

Scroll to explore

1940s-1950s

Origins: Turing to Dartmouth

The concept of AI is born. Turing Test and Dartmouth Conference laid the foundation

1943Paper

McCulloch-Pitts Neuron Model

Warren McCulloch and Walter Pitts propose the first mathematical neuron model, laying the theoretical foundation for neural networks.

#neural-networks#foundation#neuron

1950Paper

Shannon on Computer Chess

Claude Shannon publishes "Programming a Computer for Playing Chess", pioneering computer game theory.

#game-ai#chess#search

1950Paper

Turing Test

Alan Turing proposes the "Imitation Game", later known as the Turing Test, as a criterion for machine intelligence.

#turing#foundation#philosophy

1952Software

Samuel's Checkers Program

Arthur Samuel develops IBM's first checkers program, pioneering the concept of machine self-learning.

#game-ai#machine-learning#ibm

1956Software

Logic Theorist: First AI Program

Allen Newell and Herbert Simon create Logic Theorist, the first AI program capable of proving mathematical theorems.

#theorem-proving#heuristic-search#newell

1956Conference

Dartmouth Workshop

The Dartmouth Summer Research Project on Artificial Intelligence is held, marking the birth of AI as a discipline.

#foundation#conference#mccarthy

1958Milestone

Perceptron: First Neural Computer

Frank Rosenblatt invents the Perceptron Mark I, the first hardware implementation based on neural network principles.

#neural-networks#perceptron#hardware

1958Software

LISP Language Born

John McCarthy creates LISP, becoming the dominant programming language for AI research for decades.

#programming-language#lisp#mccarthy

1959Software

GPS: General Problem Solver

Newell and Simon develop GPS, the first program attempting to simulate human problem-solving processes.

#problem-solving#symbolic-ai#newell

1959Milestone

"Machine Learning" Term Coined

Arthur Samuel formally coins the term "Machine Learning", defining it as "the field of study that gives computers the ability to learn without being explicitly programmed."

#machine-learning#samuel#terminology

1960s-1970s

Early AI: Exploration & Setbacks

From ELIZA's conversation attempts to the first AI Winter caused by "Perceptrons"

1965Software

DENDRAL: First Expert System

Edward Feigenbaum and colleagues develop DENDRAL, the first knowledge-based expert system for inferring molecular structures.

#expert-systems#knowledge-engineering#chemistry

1969Milestone

SHAKEY: First Mobile Intelligent Robot

Stanford Research Institute develops SHAKEY, the first mobile robot capable of perceiving its environment and acting autonomously.

#robotics#planning#perception

1966Software

ELIZA: The First Chatbot

Joseph Weizenbaum develops ELIZA, one of the earliest natural language processing programs, capable of simulating a psychotherapist.

#nlp#chatbot#turing-test

1969Book

Perceptrons and the AI Winter

Minsky and Papert publish "Perceptrons", highlighting the limitations of single-layer neural networks, leading to the first decline in neural network research.

#neural-networks#ai-winter#history

1970Software

Prolog Language Born

Alain Colmerauer develops Prolog at University of Marseille, pioneering the logic programming paradigm.

#programming-language#logic-programming#declarative

1972Milestone

Lighthill Report and the First AI Winter

British scientist James Lighthill publishes a critical report, leading to massive funding cuts for AI research and starting the first AI Winter.

#ai-winter#funding#history

1979Milestone

Stanford Cart: Autonomous Driving Pioneer

Stanford Cart achieves autonomous navigation under Hans Moravec, traversing obstacle courses using stereo vision.

#autonomous-vehicles#computer-vision#robotics

1980s

The Rise of Expert Systems

Combining knowledge bases and inference engines, AI finds its first commercial success

1980Milestone

Japanese Fifth Generation Computer Project

Japan launches the Fifth Generation Computer Systems project, investing heavily in intelligent computers and sparking a global AI race.

#fifth-generation#japan#parallel-computing

1981Software

MYCIN: Medical Diagnosis Expert System

MYCIN expert system developed at Stanford achieves expert-level performance in diagnosing blood infections.

#expert-systems#medicine#diagnosis

1983Software

Soar Cognitive Architecture

John Laird, Allen Newell, and Paul Rosenbloom develop Soar, an important attempt at a general cognitive architecture.

#cognitive-architecture#unified-theory#cmu

1986Paper

Backpropagation Algorithm Revival

Rumelbart, Hinton, and Williams popularize the backpropagation algorithm, reviving neural network research.

#neural-networks#backpropagation#hinton

1987Milestone

Lisp Machine Crash and the Second AI Winter

Dedicated Lisp machine market collapses, AI hardware companies fail, starting the second AI Winter.

#ai-winter#lisp#expert-systems

1989Model

LeCun CNN for Handwriting Recognition

Yann LeCun at Bell Labs develops convolutional neural networks, successfully applying them to handwritten digit recognition.

#cnn#computer-vision#lecun

1989Paper

Q-learning Algorithm

Chris Watkins proposes Q-learning, becoming one of the most fundamental and widely used algorithms in reinforcement learning.

#reinforcement-learning#q-learning#decision-making

1980Software

XCON: Commercial Success of Expert Systems

XCON (R1) saves DEC millions of dollars annually, marking a huge commercial success for expert systems.

#expert-systems#commercial-ai#rules

1990s-2000s

Statistical Learning & Deep Blue

Probabilistic models become mainstream; computer defeats human chess champion

1995Paper

Support Vector Machine (SVM)

Vapnik and colleagues refine SVM theory, becoming one of the most important algorithms in machine learning.

#machine-learning#svm#statistical-learning

1997Paper

LSTM Neural Network

Hochreiter and Schmidhuber propose LSTM, solving the long-term dependency problem of recurrent neural networks.

#neural-networks#lstm#sequence-modeling

1998Model

LeNet-5 and MNIST

Yann LeCun publishes LeNet-5 architecture and MNIST dataset, becoming classic benchmarks for deep learning.

#cnn#mnist#lecun

1998Paper

PageRank Algorithm

Larry Page and Sergey Brin invent PageRank, laying the core technology foundation for Google search engine.

#search#google#algorithm

1997Milestone

Deep Blue Beats Kasparov

IBM's Deep Blue computer defeats world chess champion Garry Kasparov in a standard match.

#chess#ibm#game-ai

2001Paper

Random Forest Algorithm

Leo Breiman proposes Random Forest, becoming one of the most successful ensemble learning methods in machine learning.

#machine-learning#random-forest#ensemble-learning

2002Product

Roomba Home Robot

iRobot launches Roomba home vacuum robot, becoming the most successful consumer robot product.

#robotics#consumer-product#irobot

2006Paper

Deep Belief Network: Deep Learning Revival

Geoffrey Hinton proposes Deep Belief Networks, pioneering effective methods for training deep learning.

#deep-learning#hinton#neural-networks

2009Milestone

ImageNet Dataset

Fei-Fei Li's team releases ImageNet large-scale image dataset, becoming key infrastructure for deep learning development.

#dataset#imagenet#computer-vision

2010s

The Explosion of Deep Learning

From AlexNet to AlphaGo, big data and computing power usher in a new era of AI

15.3%

Top-5 Error Rate

AlexNet reduced error rate from 26% to 15.3%

4:1

AlphaGo vs Lee Sedol

AI defeats human Go world champion for the first time

2012Model

AlexNet: Deep Learning Breakthrough

AlexNet wins the ImageNet competition by a large margin, marking the beginning of the deep learning era.

#deep-learning#cnn#imagenet

2011Milestone

Watson Defeats Jeopardy Human Champions

IBM Watson defeats human champions in the quiz show Jeopardy, demonstrating breakthrough in natural language understanding.

#nlp#qa-system#ibm

2013Model

Word2Vec: Word Embedding Revolution

Google's Tomas Mikolov releases Word2Vec, pioneering a new era of word embedding technology.

#nlp#word-embedding#word2vec

2016Milestone

AlphaGo Defeats Lee Sedol

DeepMind's AlphaGo defeats Go world champion Lee Sedol 4-1, demonstrating the power of deep reinforcement learning.

#reinforcement-learning#go#deepmind

2014Paper

GAN: Generative Adversarial Networks

Ian Goodfellow proposes Generative Adversarial Networks, pioneering a new paradigm for deep generative models.

#generative-ai#gan#deep-learning

2014Paper

Dropout Regularization

Geoffrey Hinton and colleagues propose Dropout technique, effectively preventing neural network overfitting.

#deep-learning#regularization#hinton

2014Model

VGGNet: Very Deep Convolutional Networks

Oxford VGG team proposes VGGNet, demonstrating that stacking small convolution kernels can build deeper networks.

#cnn#deep-learning#computer-vision

2014Paper

Attention Mechanism: New Approach to Sequence Alignment

Dzmitry Bahdanau and colleagues propose attention mechanism, solving long-range dependency problem in sequence-to-sequence models.

#attention#nlp#machine-translation

2015Model

ResNet: Residual Networks Breakthrough

Microsoft Research proposes ResNet, solving deep network training difficulties through residual connections.

#cnn#deep-learning#resnet

2015Milestone

OpenAI Founded

Elon Musk, Sam Altman, and others found OpenAI, dedicated to developing safe artificial general intelligence.

#openai#organization#agi

2015Software

TensorFlow Open Sourced

Google open sources TensorFlow deep learning framework, becoming one of the most popular machine learning platforms.

#tensorflow#deep-learning#framework

2016Milestone

Microsoft Speech Recognition Reaches Human Parity

Microsoft announces its speech recognition system achieves human parity on Switchboard benchmark.

#speech-recognition#microsoft#deep-learning

2017–2019

Transformer: The Attention Revolution

The paper "Attention Is All You Need" is published, introducing the Transformer architecture and revolutionizing NLP.

In 2017, Ashish Vaswani and colleagues from Google Brain published the milestone paper "Attention Is All You Need", introducing the Transformer architecture. Transformer is based entirely on attention mechanisms, abandoning the previously mainstream recurrent neural network (RNN) and convolutional neural network (CNN) structures. Its core self-attention mechanism can process sequential data in parallel, greatly improving training efficiency. Transformer not only changed the field of natural language processing but also laid the foundation for the birth of large language models like GPT and BERT.

#transformer#attention#nlp#google

Learn More

Transformer Architecture

Encoder

Self-Attention → Feed Forward

Decoder

Masked Attention → Cross-Attention → FFN

Parallel Processing

Self-Attention

Position Encoding

Scalability

2017Model

AlphaGo Zero: No Human Knowledge Required

DeepMind releases AlphaGo Zero, learning only through self-play and surpassing all Go AI.

#reinforcement-learning#go#deepmind

2017Software

PyTorch Released

Facebook releases PyTorch deep learning framework, winning researchers over with its dynamic computation graph.

#pytorch#deep-learning#framework

2018Model

BERT: Bidirectional Language Understanding

Google proposes BERT, achieving breakthrough language understanding through bidirectional Transformers.

#nlp#bert#transformer

2018Model

GPT-1: Generative Pre-Training

OpenAI releases GPT-1, first demonstrating the potential of generative pre-training on language understanding tasks.

#nlp#gpt#transformer

2018Model

StyleGAN: High-Quality Image Generation

NVIDIA proposes StyleGAN, generating unprecedented high-quality face images.

#generative-ai#gan#stylegan

2019Model

GPT-2: Better Language Models

OpenAI releases GPT-2 in stages due to misuse concerns, demonstrating emergent capabilities of large models.

#nlp#gpt#llm

2019Milestone

OpenAI Five: Dota2 AI Champion

OpenAI Five defeats Dota2 world champion team, demonstrating AI capability in complex team games.

#reinforcement-learning#game-ai#dota2

2020s

Large Language Models & Generative AI

GPT, Claude, Gemini, and other LLMs demonstrate remarkable capabilities

2020Model

GPT-3: Scale is All You Need

OpenAI releases GPT-3 with 175 billion parameters, demonstrating emergent capabilities of large-scale language models.

#nlp#gpt#llm

2020Model

AlphaFold2: Protein Folding Breakthrough

DeepMind's AlphaFold2 achieves experimental accuracy in CASP14, solving a 50-year biology challenge.

#biology#protein-folding#deepmind

2021Model

DALL·E: Text to Image

OpenAI releases DALL·E, first achieving high-quality image generation from text descriptions.

#generative-ai#text-to-image#dall-e

2021Model

CLIP: Connecting Vision and Language

OpenAI releases CLIP, achieving powerful vision-language understanding through contrastive learning.

#multimodal#clip#vision-language

2021Product

GitHub Copilot: AI Programming Assistant

GitHub and OpenAI release Copilot, pioneering a new era of AI-assisted programming.

#code-generation#copilot#openai

2022Model

Stable Diffusion: Open Source Image Generation

Stability AI releases open-source Stable Diffusion, sparking the text-to-image generation revolution.

#generative-ai#text-to-image#stable-diffusion

2022Product

Midjourney: Artistic Image Generation

Midjourney releases, winning designers' favor with unique artistic style generation capabilities.

#generative-ai#text-to-image#midjourney

2022Model

DALL·E 2: High-Resolution Image Generation

OpenAI releases DALL·E 2, generating higher quality, more realistic images.

#generative-ai#text-to-image#dall-e

2022Paper

InstructGPT: RLHF Breakthrough

OpenAI releases InstructGPT, demonstrating alignment of language models through Reinforcement Learning from Human Feedback.

#rlhf#alignment#instructgpt

2022Product

ChatGPT Launch

OpenAI releases ChatGPT, which takes the world by storm with its amazing conversational abilities, starting the Generative AI boom.

#llm#chatgpt#openai

2023Model

LLaMA: Open Source LLM Wave

Meta releases LLaMA models, opening a new era of open-source large language models.

#llm#llama#meta

2023Model

Claude: Safety-First AI Assistant

Anthropic releases Claude, emphasizing AI safety and helpfulness.

#llm#claude#anthropic

2023Model

Llama 2: Commercial Open-Source LLM

Meta releases Llama 2, the first commercially usable open-source large language model.

#llm#llama#meta

2023Model

Mistral 7B: Small but Powerful

Mistral AI releases Mistral 7B, proving small models can achieve powerful performance.

#llm#mistral#efficient-ai

2023Model

Gemini: Google Multimodal Model

Google releases Gemini multimodal LLM, natively supporting text, image, audio, and video.

#llm#gemini#google

2023Milestone

Waymo Fully Driverless Commercial License

Waymo receives 24/7 fully driverless commercial operation permit in San Francisco.

#autonomous-driving#waymo#self-driving

2023Model

GPT-4 Launch

OpenAI releases GPT-4, demonstrating powerful multimodal and reasoning capabilities, sparking global discussion about AGI.

#llm#gpt#openai

2024Model

Sora: Video Generation Breakthrough

OpenAI releases Sora video generation model, capable of generating up to one minute of high-quality video.

#generative-ai#text-to-video#sora

2024Model

Claude 3 Series Released

Anthropic releases Claude 3 series, with Opus version surpassing GPT-4 on multiple tests.

#llm#claude#anthropic

2024Model

GPT-4o: Full Modal Real-time Interaction

OpenAI releases GPT-4o, achieving millisecond-response multimodal real-time interaction.

#llm#gpt#openai

2024Model

DeepSeek V2: Chinese Open-Source LLM

DeepSeek releases V2 model, achieving high performance at extremely low cost, leading new direction in open-source models.

#llm#deepseek#open-source

2024Product

Apple Intelligence: On-Device AI

Apple releases Apple Intelligence, deeply integrating AI capabilities on-device.

#on-device-ai#apple#privacy

2024Model

OpenAI o1: Reasoning Breakthrough

OpenAI releases o1 model, demonstrating powerful reasoning and problem-solving capabilities.

#llm#reasoning#openai

2025Model

DeepSeek R1: Open-Source Reasoning Model

DeepSeek releases R1 reasoning model, matching o1 performance and fully open-sourced.

#llm#reasoning#deepseek

2025Model

Claude 3.5/4 Series

Anthropic continues updating Claude series, leading in coding and reasoning capabilities.

#llm#claude#anthropic

2025Model

GPT-4.5/o3: Next Generation Models

OpenAI releases GPT-4.5 and o3 series, continuously pushing LLM capability boundaries.

#llm#gpt#openai

Future

Future: Promises & Controversies

AGI possibilities, safety challenges, and social implications

🚀 Opportunities

✓Accelerated scientific research
✓Personalized education and healthcare
✓Creativity augmentation

⚠️ Challenges

!Job displacement and inequality
!Bias and fairness
!Safety and control concerns