AI History in Papers
한국어

AI History in Papers

The history of AI research from 1943 to today, followed through its original papers

Key point

  • Every dot on the timeline is an original paper, with a link to the source, one sentence quoted from it with a page number, and one or two lines on what changed.
  • Lineage lines are solid only where a citation is confirmed. Influences stated in textbooks without a confirmed citation are dotted.
  • The selection rules were registered before the list was made (see Method and limits). Priority disputes show both sides' bibliographies and make no ruling.
191original papers1108confirmed citation lines1943–2025first-public years
Ch. 1 Origins 1943–56 Ch. 2 Symbols and perceptrons 1956–74 Ch. 3 Knowledge systems and connectionism 1974–93 Ch. 4 Statistical learning 1990–2011 Ch. 5 Deep learning 2012–17 Ch. 6 Pretraining and generation 2017–22 Ch. 7 Agents 2022– Neural network learning Reinforcement learning and games Language and speech Vision and generative models Search, planning and knowledge Probability, statistics and origins Scale, alignment and safety Agents 1950 1960 1970 1980 1990 2000 2010 2012 2014 2016 2018 2020 2022 2024 2026
One dot per paper. Horizontal position is the first public year (the axis gives the 70 years before 2012 and the 14 years after it similar widths); rows are theme groups. Hover a dot for its card.
  1. 1943 — McCulloch & Pitts, A logical calculus of the ideas immanent in nervous activity
  2. 1948 — Shannon, A Mathematical Theory of Communication
  3. 1948 — Wiener, Cybernetics: or Control and Communication in the Animal and the Machine
  4. 1949 — Shannon, Programming a computer for playing chess
  5. 1949 — Hebb, The Organization of Behavior
  6. 1950 — Turing, Computing Machinery and Intelligence
  7. 1956 — Chomsky, Three models for the description of language
  8. 1957 — Bellman, Dynamic Programming
  9. 1958 — McCarthy, Programs with Common Sense
  10. 1958 — Rosenblatt, The perceptron: A probabilistic model for information storage and organization in the brain
  11. 1959 — Newell et al., Report on a general problem-solving program
  12. 1959 — Samuel, Some Studies in Machine Learning Using the Game of Checkers
  13. 1960 — McCarthy, Recursive functions of symbolic expressions and their computation by machine, Part I
  14. 1960 — Widrow & Hoff, Adaptive switching circuits
  15. 1961 — Minsky, Steps toward Artificial Intelligence
  16. 1965 — Robinson, A Machine-Oriented Logic Based on the Resolution Principle
  17. 1965 — Zadeh, Fuzzy sets
  18. 1966 — Weizenbaum, ELIZA—a computer program for the study of natural language communication between man and machine
  19. 1968 — Hart et al., A Formal Basis for the Heuristic Determination of Minimum Cost Paths
  20. 1969 — McCarthy & Hayes, Some Philosophical Problems from the Standpoint of Artificial Intelligence
  21. 1969 — Minsky & Papert, Perceptrons: An Introduction to Computational Geometry
  22. 1970 — Fikes & Nilsson, STRIPS: A new approach to the application of theorem proving to problem solving
  23. 1971 — Winograd, Understanding natural language
  24. 1973 — Lighthill, Artificial Intelligence: A General Survey
  25. 1974 — Minsky, A Framework for Representing Knowledge
  26. 1974 — Werbos, Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences
  27. 1975 — Holland, Adaptation in Natural and Artificial Systems
  28. 1975 — Newell & Simon, Computer science as empirical inquiry: symbols and search
  29. 1976 — Dempster et al., Maximum Likelihood from Incomplete Data Via the EM Algorithm
  30. 1977 — Schank & Abelson, Scripts, Plans, Goals and Understanding
  31. 1978 — Buchanan & Feigenbaum, Dendral and Meta-Dendral: Their applications dimension
  32. 1980 — Fukushima, Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position
  33. 1980 — McDermott, R1: A rule-based configurer of computer systems
  34. 1980 — Searle, Minds, brains, and programs
  35. 1982 — Hopfield, Neural networks and physical systems with emergent collective computational abilities
  36. 1982 — Marr, Vision: A Computational Investigation into the Human Representation and Processing of Visual Information
  37. 1982 — Werbos, Applications of advances in nonlinear sensitivity analysis
  38. 1983 — Kirkpatrick et al., Optimization by Simulated Annealing
  39. 1983 — Barto et al., Neuronlike adaptive elements that can solve difficult learning control problems
  40. 1984 — Valiant, A theory of the learnable
  41. 1984 — Geman & Geman, Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
  42. 1985 — Ackley et al., A learning algorithm for Boltzmann machines
  43. 1986 — Quinlan, Induction of decision trees
  44. 1986 — Rumelhart et al., Learning representations by back-propagating errors
  45. 1987 — Waibel et al., Phoneme recognition using time-delay neural networks
  46. 1988 — Elman, Finding Structure in Time
  47. 1988 — Hornik et al., Multilayer feedforward networks are universal approximators
  48. 1988 — Pearl, Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
  49. 1988 — Sutton, Learning to predict by the methods of temporal differences
  50. 1989 — Rabiner, A tutorial on hidden Markov models and selected applications in speech recognition
  51. 1989 — Watkins, Learning from Delayed Rewards
  52. 1989 — Schapire, The strength of weak learnability
  53. 1989 — Cybenko, Approximation by superpositions of a sigmoidal function
  54. 1989 — LeCun et al., Backpropagation Applied to Handwritten Zip Code Recognition
  55. 1990 — Brown et al., A statistical approach to machine translation
  56. 1991 — Jacobs et al., Adaptive Mixtures of Local Experts
  57. 1991 — Hochreiter, Untersuchungen zu dynamischen neuronalen Netzen
  58. 1992 — Watkins & Dayan, Q-learning
  59. 1992 — Williams, Simple statistical gradient-following algorithms for connectionist reinforcement learning
  60. 1992 — Boser et al., A training algorithm for optimal margin classifiers
  61. 1993 — Quinlan, C4.5: Programs for Machine Learning
  62. 1994 — Bengio et al., Learning long-term dependencies with gradient descent is difficult
  63. 1994 — Breiman, Bagging predictors
  64. 1995 — Freund & Schapire, A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting
  65. 1995 — Tesauro, Temporal difference learning and TD-Gammon
  66. 1995 — Hochreiter & Schmidhuber, Long Short-Term Memory
  67. 1995 — Russell & Norvig, Artificial Intelligence: A Modern Approach
  68. 1995 — Cortes & Vapnik, Support-vector networks
  69. 1996 — Tibshirani, Regression Shrinkage and Selection Via the Lasso
  70. 1997 — Schuster & Paliwal, Bidirectional recurrent neural networks
  71. 1998 — Kaelbling et al., Planning and acting in partially observable stochastic domains
  72. 1998 — Sutton & Barto, Reinforcement Learning: An Introduction
  73. 1998 — LeCun et al., Gradient-based learning applied to document recognition
  74. 1999 — Sutton et al., Policy Gradient Methods for Reinforcement Learning with Function Approximation
  75. 2000 — Hinton, Training Products of Experts by Minimizing Contrastive Divergence
  76. 2001 — Lafferty et al., Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
  77. 2001 — Breiman, Random Forests
  78. 2002 — Campbell et al., Deep Blue
  79. 2003 — Blei et al., Latent Dirichlet Allocation
  80. 2003 — Bengio et al., A Neural Probabilistic Language Model
  81. 2006 — Coulom, Efficient Selectivity and Backup Operators in Monte-Carlo Tree Search
  82. 2006 — Hinton et al., A Fast Learning Algorithm for Deep Belief Nets
  83. 2006 — Hinton & Salakhutdinov, Reducing the Dimensionality of Data with Neural Networks
  84. 2006 — Kocsis & Szepesvári, Bandit Based Monte-Carlo Planning
  85. 2006 — Thrun et al., Stanley: The robot that won the DARPA Grand Challenge
  86. 2006 — Bengio et al., Greedy Layer-Wise Training of Deep Networks
  87. 2007 — Schaeffer et al., Checkers Is Solved
  88. 2009 — Deng et al., ImageNet: A large-scale hierarchical image database
  89. 2009 — Raina et al., Large-scale deep unsupervised learning using graphics processors
  90. 2010 — Ciresan et al., Deep, Big, Simple Neural Nets for Handwritten Digit Recognition
  91. 2010 — Glorot & Bengio, Understanding the difficulty of training deep feedforward neural networks
  92. 2010 — Nair & Hinton, Rectified Linear Units Improve Restricted Boltzmann Machines
  93. 2010 — Mikolov et al., Recurrent neural network based language model
  94. 2011 — Collobert et al., Natural Language Processing (almost) from Scratch
  95. 2011 — Le et al., Building high-level features using large scale unsupervised learning
  96. 2012 — Ciresan et al., Multi-column deep neural networks for image classification
  97. 2012 — Hinton et al., Deep Neural Networks for Acoustic Modeling in Speech Recognition
  98. 2012 — Krizhevsky et al., ImageNet Classification with Deep Convolutional Neural Networks
  99. 2013 — Mikolov et al., Efficient Estimation of Word Representations in Vector Space
  100. 2013 — Graves, Generating Sequences With Recurrent Neural Networks
  101. 2013 — Mikolov et al., Distributed Representations of Words and Phrases and their Compositionality
  102. 2013 — Girshick et al., Rich feature hierarchies for accurate object detection and semantic segmentation
  103. 2013 — Kingma & Welling, Auto-Encoding Variational Bayes
  104. 2013 — Mnih et al., Playing Atari with Deep Reinforcement Learning
  105. 2013 — Szegedy et al., Intriguing properties of neural networks
  106. 2014 — Schmidhuber, Deep learning in neural networks: An overview
  107. 2014 — Cho et al., Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
  108. 2014 — Goodfellow et al., Generative Adversarial Nets
  109. 2014 — Srivastava et al., Dropout: A Simple Way to Prevent Neural Networks from Overfitting
  110. 2014 — Bahdanau et al., Neural Machine Translation by Jointly Learning to Align and Translate
  111. 2014 — Simonyan & Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition
  112. 2014 — Sutskever et al., Sequence to Sequence Learning with Neural Networks
  113. 2014 — Szegedy et al., Going Deeper with Convolutions
  114. 2014 — Graves et al., Neural Turing Machines
  115. 2014 — Pennington et al., GloVe: Global Vectors for Word Representation
  116. 2014 — Weston et al., Memory Networks
  117. 2014 — Kingma & Ba, Adam: A Method for Stochastic Optimization
  118. 2015 — Ioffe & Szegedy, Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
  119. 2015 — Mnih et al., Human-level control through deep reinforcement learning
  120. 2015 — Schulman et al., Trust Region Policy Optimization
  121. 2015 — Hinton et al., Distilling the Knowledge in a Neural Network
  122. 2015 — Sohl-Dickstein et al., Deep Unsupervised Learning using Nonequilibrium Thermodynamics
  123. 2015 — LeCun et al., Deep learning
  124. 2015 — Luong et al., Effective Approaches to Attention-based Neural Machine Translation
  125. 2015 — Sennrich et al., Neural Machine Translation of Rare Words with Subword Units
  126. 2015 — Lillicrap et al., Continuous control with deep reinforcement learning
  127. 2015 — Radford et al., Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
  128. 2015 — He et al., Deep Residual Learning for Image Recognition
  129. 2016 — Silver et al., Mastering the game of Go with deep neural networks and tree search
  130. 2016 — Mnih et al., Asynchronous Methods for Deep Reinforcement Learning
  131. 2016 — Amodei et al., Concrete Problems in AI Safety
  132. 2016 — Ba et al., Layer Normalization
  133. 2016 — van den Oord et al., WaveNet: A Generative Model for Raw Audio
  134. 2016 — Wu et al., Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
  135. 2017 — Shazeer et al., Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
  136. 2017 — Jouppi et al., In-Datacenter Performance Analysis of a Tensor Processing Unit
  137. 2017 — Christiano et al., Deep reinforcement learning from human preferences
  138. 2017 — Vaswani et al., Attention Is All You Need
  139. 2017 — Schulman et al., Proximal Policy Optimization Algorithms
  140. 2017 — Silver et al., Mastering the game of Go without human knowledge
  141. 2017 — Hestness et al., Deep Learning Scaling is Predictable, Empirically
  142. 2018 — Howard & Ruder, Universal Language Model Fine-tuning for Text Classification
  143. 2018 — Peters et al., Deep contextualized word representations
  144. 2018 — Radford et al., Improving Language Understanding by Generative Pre-Training
  145. 2018 — Devlin et al., BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
  146. 2018 — Silver et al., A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
  147. 2019 — Radford et al., Language Models are Unsupervised Multitask Learners
  148. 2019 — Raffel et al., Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
  149. 2019 — Schrittwieser et al., Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
  150. 2020 — Kaplan et al., Scaling Laws for Neural Language Models
  151. 2020 — Brown et al., Language Models are Few-Shot Learners
  152. 2020 — Lewis et al., Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
  153. 2020 — Stiennon et al., Learning to summarize from human feedback
  154. 2020 — Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
  155. 2021 — Fedus et al., Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
  156. 2021 — Radford et al., Learning Transferable Visual Models From Natural Language Supervision
  157. 2021 — Bender et al., On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?
  158. 2021 — Chen et al., Evaluating Large Language Models Trained on Code
  159. 2021 — Jumper et al., Highly accurate protein structure prediction with AlphaFold
  160. 2021 — Bommasani et al., On the Opportunities and Risks of Foundation Models
  161. 2021 — Nakano et al., WebGPT: Browser-assisted question-answering with human feedback
  162. 2022 — Wei et al., Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  163. 2022 — Hoffmann et al., Training Compute-Optimal Large Language Models
  164. 2022 — Ouyang et al., Training language models to follow instructions with human feedback
  165. 2022 — Wang et al., Self-Consistency Improves Chain of Thought Reasoning in Language Models
  166. 2022 — Ahn et al., Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
  167. 2022 — Chowdhery et al., PaLM: Scaling Language Modeling with Pathways
  168. 2022 — Kojima et al., Large Language Models are Zero-Shot Reasoners
  169. 2022 — Wei et al., Emergent Abilities of Large Language Models
  170. 2022 — Huang et al., Inner Monologue: Embodied Reasoning through Planning with Language Models
  171. 2022 — Yao et al., ReAct: Synergizing Reasoning and Acting in Language Models
  172. 2023 — Greshake et al., Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection
  173. 2023 — Schick et al., Toolformer: Language Models Can Teach Themselves to Use Tools
  174. 2023 — Touvron et al., LLaMA: Open and Efficient Foundation Language Models
  175. 2023 — OpenAI, GPT-4 Technical Report
  176. 2023 — Shen et al., HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
  177. 2023 — Shinn et al., Reflexion: Language Agents with Verbal Reinforcement Learning
  178. 2023 — Park et al., Generative Agents: Interactive Simulacra of Human Behavior
  179. 2023 — Wang et al., Voyager: An Open-Ended Embodied Agent with Large Language Models
  180. 2023 — Yao et al., Tree of Thoughts: Deliberate Problem Solving with Large Language Models
  181. 2023 — Zhou et al., WebArena: A Realistic Web Environment for Building Autonomous Agents
  182. 2023 — Hong et al., MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
  183. 2023 — Liu et al., AgentBench: Evaluating LLMs as Agents
  184. 2023 — Jimenez et al., SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
  185. 2023 — Packer et al., MemGPT: Towards LLMs as Operating Systems
  186. 2023 — Mialon et al., GAIA: a benchmark for General AI Assistants
  187. 2024 — Xie et al., OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
  188. 2024 — Yang et al., SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
  189. 2024 — Yao et al., τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
  190. 2024 — Kapoor et al., AI Agents That Matter
  191. 2025 — DeepSeek-AI, DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Eras

Origins: cybernetics and computation (1943–1956)Can brain, machine and information be written in one language?7 papersSymbolic AI and the first neural networks (1956–1974)Can intelligence be written as search and logic, and how much can a perceptron learn?18 papersKnowledge systems and the return of connectionism (1974–1993)Can expert knowledge be written as rules, and can multilayer networks be trained?28 papersStatistical learning (1990–2011)Can generalization from data be guaranteed?41 papersDeep learning (2012–2017)Deep networks, GPUs and large datasets arrive together.46 papersPretraining and generative models (2017–2022)Scale, self-supervision and alignment.29 papersAgents (2022– )Models use tools and act.22 papers

What this site does

The history of AI research has been written in many places: the Wikipedia timelines, the Computer History Museum's timeline, the history chapters of textbooks, one researcher's long annotated history. This site does not replace them. It adds one thing: every point is tied to an original paper. Dates are first-public dates rather than popularly quoted years, "what changed" is written from the paper's own sentences, and lines of influence distinguish confirmed citations from unconfirmed attributions.

Numbers only support the narrative. Citation counts appear small on each card with the date they were read, and there is no ranking. The page on winters and booms re-examines the two conventional winters (around 1974 and around 1987) against paper-title counts, conference attendance and public funding records.