Research history

Mathematician Norbert Wiener is widely considered the founder of cybernetics, the study of regulatory systems. His 1948 book Cybernetics treated feedback formally and framed control and communication as one problem for animals and machines.

Alan Turing (1912–1954) was a British mathematician whose 1936 paper defined computation with an abstract machine. His 1950 paper proposed the imitation game, now called the Turing test, and he is counted among the founders of artificial intelligence.

American mathematician and electrical engineer whose 1948 paper "A Mathematical Theory of Communication" gave communication engineering a mathematical foundation.

John McCarthy (1927–2011) was an American computer scientist. His 1955 proposal is credited with coining the term “artificial intelligence”, he organised the 1956 Dartmouth workshop, and he created the LISP programming language.

Alan Turing's 1936 paper defined computable numbers through an abstract machine, now called the Turing machine, and showed that no general method can decide whether any given mathematical statement is provable.

An artificial neural network is a computing system of many simple connected units whose behaviour is set by connection weights. Its roots lie in 1943 models of neurons, and its learning versions began with Frank Rosenblatt's perceptron in 1958.

In 1943 Warren McCulloch and Walter Pitts described a simplified model of a neuron and showed that networks of such units can carry out logical reasoning, an early link between brain models and computation.

ENIAC, announced on 14 February 1946, is described by the University of Pennsylvania as the first general-purpose electronic computer: a 30-ton machine with 18,000 vacuum tubes, built to produce military firing tables.

Geoffrey Hinton is a computer scientist born in London whose research on neural networks, including back-propagation and deep convolutional networks, is central to deep learning. He shared the 2018 ACM A.M. Turing Award and the 2024 Nobel Prize in Physics.

C. E. Shannon's 1948 paper set out a mathematical framework for communication, defining information in bits, channel capacity, and entropy, and treating messages statistically while setting aside semantic meaning.

A framework that treats information as a measurable quantity to be encoded and transmitted, founded in Shannon's 1948 paper; it gave the field the bit, entropy and channel capacity.

Marc Raibert (born 1949) is a robotics researcher who built the first self-balancing hopping robots, founded Boston Dynamics, and is now executive director of the Boston Dynamics AI Institute.

The Turing test is Alan Turing's 1950 proposal for replacing the question "Can machines think?" with a game: a machine passes if an interrogator cannot tell its written answers apart from a human's.

In 1950 Alan Turing replaced the question "Can machines think?" with the imitation game, now called the Turing test, and answered the main objections to machine intelligence one by one.

Logic Theorist, completed in 1956 by Allen Newell, Herbert A. Simon and Cliff Shaw, was the first program deliberately built to perform automated reasoning; it proved dozens of theorems from Principia Mathematica and introduced heuristics, list processing and reasoning as search.

In 1958 Frank Rosenblatt described the perceptron, a model whose connection strengths change with training so that it learns to classify stimuli by trial and error. It became the starting point for trainable neural networks.

Lisp is a programming language invented by John McCarthy at MIT in 1958, whose programs are themselves lists. It became the favoured language of early artificial intelligence research and introduced automatic garbage collection.

John McCarthy's 1959 paper proposing the advice taker, a program that would solve problems by manipulating sentences in formal languages and act when it deduced an imperative — an early statement of logical AI.

A. L. Samuel's 1959 paper reporting experiments in which a checkers program, given only the rules of the game and an imperfect set of parameters, learned in 8 to 10 hours of playing time to beat the person who wrote it.

The Mark I Perceptron was the first hardware implementation of the perceptron algorithm: a custom analog computer built at Cornell Aeronautical Laboratory for image recognition, with photocells as input and motor-driven potentiometers as learnable weights.

Yann LeCun is a French computer scientist known for convolutional neural networks, which learn to recognise images directly from pixels. He co-authored the 1998 paper that introduced LeNet-5 and shared the 2018 ACM A.M. Turing Award.

Unimate was an industrial robotic arm designed by George C. Devol and Joseph F. Engelberger and built by Unimation; in 1961 it became the first robot installed on an assembly line, unloading a die-casting press at a General Motors plant.

Marvin Minsky's 1961 survey in Proceedings of the IRE organized heuristic programming into five problem areas — search, pattern recognition, learning, planning and induction — and argued that machines would need to build models of their environments.

DENDRAL was a Stanford artificial intelligence project begun in 1964 that identified unknown organic molecules from mass spectra using chemical knowledge. It is widely described as the first expert system and the ancestor of later systems such as MYCIN and PROSPECTOR.

Canadian computer scientist whose work on sequence models, word embeddings and attention helped lay foundations for modern language models. He shared the 2018 ACM A.M. Turing Award with Geoffrey Hinton and Yann LeCun.

Shakey was a mobile robot built at SRI International between 1966 and 1972, the first AI-based mobile robot that could perceive and reason about its surroundings and act without step-by-step instructions. Its project produced the A* search algorithm and the STRIPS planning system.

A 1969 book by Marvin Minsky and Seymour Papert that proved mathematical limits on what single-layer perceptrons can compute — notably that they cannot compute parity or connectedness — and has long been credited with steering research away from neural networks.

AI winter is a period of sharply reduced funding and interest in artificial intelligence research. The field's hype cycles have produced two such downturns, usually dated to about 1974–1980 and 1987–2000.

A recurrent neural network introduced by John Hopfield in 1982 that stores patterns as minima of an energy function and recalls them from partial or distorted input, bringing statistical mechanics into the study of neural networks.

A 1982 paper by John J. Hopfield showing that large networks of simple neuron-like units can collectively store and recall memories. It became known as the Hopfield network, and the work was recognised by the 2024 Nobel Prize in Physics.

A 1982 paper by John J. Hopfield arguing that useful computation can emerge as a collective property of many simple, equivalent units, and describing a model whose collective behaviour gives content-addressable memory, generalisation and error correction.

Dario Amodei is the co-founder and CEO of Anthropic. He previously worked at OpenAI, where he was Vice President of Research and co-authored the GPT-2 paper and led the research for GPT-3.

Autonomous driving is the operation of a road vehicle by a computer system instead of a person. Its six automation levels, its two main technical traditions and its long line of research vehicles define how the field is discussed today.

Israeli-Canadian computer scientist born in 1986 who co-built AlexNet, co-founded OpenAI and served as its chief scientist, and later founded Safe Superintelligence Inc.

A two-volume MIT Press work by Rumelhart, McClelland and the PDP Research Group, published in 1986, that set out the parallel distributed processing framework — networks of simple units with hidden layers and sigmoid activation — and became a landmark of connectionism.

Backpropagation is the procedure that lets a multilayer neural network learn by sending output errors backward to adjust its weights, using gradient descent. A 1986 Nature paper made it widely known, and it is standard in most neural networks today.

In a 1986 Nature paper, Rumelhart, Hinton and Williams showed that backpropagation lets hidden units in a multilayer network learn useful internal features of a task. The method is standard in most neural networks today.

Imitation learning trains an agent by supervised learning on expert demonstrations rather than by trial and error. Its simplest form is behaviour cloning, and its best-known weakness is distribution shift.

ALVINN was a 1988 Carnegie Mellon neural network that learned to steer the Navlab van by watching a human driver, an early demonstration of learned end-to-end driving.

Richard Sutton's 1988 paper introduced temporal-difference learning, a family of incremental prediction methods that assign credit from changes between successive predictions and are shown to be more economical and accurate than conventional supervised approaches on most prediction problems.

A convolutional neural network applies the same small pattern detectors across a whole image, so a feature is found wherever it appears. Trained with backpropagation in 1989 and refined as LeNet-5 in 1998, it became standard in computer vision.

Rodney Brooks's 1990 paper argued for building artificial intelligence from the ground up, with ongoing physical interaction with the environment as the main constraint on design, rather than through the abstract manipulation of symbols that had dominated the field for roughly thirty years.

Published in 1992, this paper by Boser, Guyon and Vapnik presents a training algorithm that maximizes the margin between training patterns and the decision boundary, with experiments on optical character recognition.[^1]

Honda's P2, announced in December 1996, was presented as the world's first bipedal humanoid walking robot: a 1,820 mm, 210 kg machine that walked, climbed stairs and pushed carts without cables, and led to ASIMO.

IBM's chess computer that in May 1997 became the first computer system to defeat a reigning world chess champion in a match under standard tournament time controls.

A small NASA rover that landed on Mars on July 4, 1997 with the Mars Pathfinder lander and became the first wheeled vehicle operated on another planet.

Published in 1997 by Sepp Hochreiter and Jürgen Schmidhuber, the LSTM paper introduced a memory cell with gates and a "constant error carousel" that lets recurrent networks learn across long time gaps. It shaped the recurrent models that came before Transformers.

A 1998 conference paper by Sergey Brin and Lawrence Page describing PageRank, an algorithm that ranks web pages by the number and quality of links pointing to them, and the search engine built on that algorithm.

Sony's ERS-110, announced in May 1999, was the first AIBO: a four-legged autonomous entertainment robot that learned from interaction with its owner and expressed emotions.

ASIMO is a humanoid robot Honda announced in Tokyo on 20 November 2000, built to walk, climb stairs and work in human-sized spaces; successive versions ran faster and handled objects until its last regular demonstration in 2022.

A generative probabilistic model that represents each document as a finite mixture over topics, fitted with variational approximate inference and EM, and evaluated on document modelling, text classification and collaborative filtering.

Stanley is the Stanford University robotic vehicle that won DARPA's 2005 Grand Challenge, completing a 132-mile desert course autonomously in 6 hours 53 minutes and claiming the $2 million prize.

A 2006 paper by Hinton, Osindero and Teh that uses complementary priors and a greedy layer-by-layer algorithm to train deep belief nets, then fine-tunes them; the resulting three-hidden-layer model both modelled and classified handwritten digits well.

Siri, Inc. was a startup spun out of SRI International in 2007 to turn decades of AI research — notably the CALO project — into a consumer virtual personal assistant; Apple acquired it in 2010.

A 2009 ICML paper by Rajat Raina, Anand Madhavan and Andrew Y. Ng that used graphics processors to train deep belief networks and sparse coding models many times faster than multicore CPUs.

CIFAR-10 is a labelled image dataset of 60000 32×32 colour images in 10 mutually exclusive classes, split into 50000 training and 10000 test images; it was created by Krizhevsky, Nair and Hinton for training and comparing image classifiers.

The 2009 paper introduced ImageNet, a large, cleanly labelled image database organised by the WordNet hierarchy. It later became the basis of an annual benchmark, the ImageNet Large Scale Visual Recognition Challenge.

A 2010 paper by Xavier Glorot and Yoshua Bengio on why deep feedforward networks are hard to train: sigmoid saturation under random initialization and layer Jacobians whose singular values sit far from 1, plus a new initialization scheme that converges substantially faster.

A London-headquartered AI research organisation formed in 2023 by merging DeepMind with Google's Brain team, known for AlphaGo, AlphaFold and the Gemini model family.

A CVPR 2011 Best Paper that recovers 3D body joint positions from a single depth image without temporal information, by classifying pixels into body parts, and runs at 200 frames per second on consumer hardware.

IBM Watson is a question-answering computer system built by an IBM research team led by David Ferrucci; in February 2011 it defeated the two foremost all-time champions of the televised quiz show Jeopardy!

A 2011 paper by Glorot, Bordes and Bengio reporting that deep networks built from rectifying neurons match or beat tanh networks and reach their best performance without unsupervised pre-training on large labelled supervised tasks.

A free online version of Stanford's artificial intelligence course, taught by Sebastian Thrun and Peter Norvig, which started on October 10, 2011 and drew 160,000 students from over 190 countries.

The 2012 NeurIPS paper that introduced AlexNet, a large convolutional network trained on GPUs. It won the ILSVRC-2012 image classification competition by a wide margin and is widely seen as a turning point for deep learning in computer vision.

Atlas was a humanoid disaster-response robot built by Boston Dynamics for DARPA, unveiled on 11 July 2013 and handed to teams in the DARPA Robotics Challenge; its hydraulic line was retired in 2024.

Meta AI is the AI research and model-building organisation of Meta Platforms. Founded in late 2013 as Facebook AI Research, it produced PyTorch and the Llama and Muse model families, and it is separate from the Meta AI assistant product.

Attention scores how important each element of a sequence is relative to the others and turns those scores into weights, letting a model read any part of its input directly. It is the operation at the core of the transformer.

A generative adversarial network pits a generator that makes fake data against a discriminator that judges it, both trained in a zero-sum game; proposed by Ian Goodfellow and colleagues in 2014, it led to a series of architectures for generating images.

Goodfellow, Bengio, and co-authors proposed a generative-model framework with adversarially trained generator and discriminator.

Submitted to arXiv in September 2014 by Dzmitry Bahdanau, Kyunghyun Cho and Yoshua Bengio, this paper let a translation model look back at the whole source sentence while producing each word. This mechanism, called attention, later became a basis for Transformer models.

A 2014 study showing that stacking many small 3×3 convolution filters, up to 16–19 weight layers, improves large-scale image recognition; its models took first place in localisation and second in classification at ImageNet Challenge 2014 and were released publicly.

Ilya Sutskever, Oriol Vinyals and Quoc Le's 2014 paper showed that two LSTMs, one reading the input and one writing the output, could translate sentences and outperform a phrase-based statistical baseline. It set the encoder–decoder pattern behind later translation models.

The 2014 paper that introduced the Inception architecture and its 22-layer GoogLeNet network, which took first place in the ImageNet 2014 classification and detection challenge with a 6.67% top-5 error.

GloVe is a 2014 method that learns word vectors by factorising global word co-occurrence counts. Released together with pre-trained vectors, it became, alongside word2vec, one of the two standard ways of producing static word embeddings.

Amazon Echo is an always-on, nine-inch cylindrical speaker that answers to the name Alexa. Launched in November 2014 by invitation, it put a voice assistant into an ordinary room rather than a phone.

Adam is a stochastic optimisation algorithm from a 2014 paper by Diederik P. Kingma and Jimmy Ba that sets per-parameter step sizes from running estimates of the gradient's first and second moments, needing little memory and little hyper-parameter tuning.

A 2015 Nature review by Yann LeCun, Yoshua Bengio and Geoffrey Hinton explaining deep learning: models of multiple processing layers that learn layered representations of data, and the results this produced in speech, vision and other fields.

Google's first-generation Tensor Processing Unit is a custom chip built to speed up neural-network inference in its data centers. Deployed in 2015 and revealed in 2016, it showed that dedicated accelerators could beat CPUs and GPUs on cost-performance.

A residual connection adds a block's input to its output, so the block learns only the change it needs to make. This makes very deep networks trainable, and it is a standard part of today's transformers.

A 2015 Nature paper by Mnih, Kavukcuoglu, Silver and colleagues showed that one deep Q-network agent, given only screen pixels and the game score, learned to play 49 Atari 2600 games at a level comparable to a professional human tester.

TensorFlow is Google's second-generation machine learning system, released as an open-source library under the Apache 2.0 license on 9 November 2015, succeeding the earlier DistBelief system.

AlphaGo is a Go-playing program from DeepMind that combined deep neural networks with Monte Carlo tree search. In 2016 it beat top professional Lee Sedol in a widely watched match, a landmark for learned systems.

A 2016 paper by Barret Zoph and Quoc V. Le that uses a recurrent network trained with reinforcement learning to design neural network architectures automatically, matching human-designed models on CIFAR-10 and beating LSTM on Penn Treebank.

PyTorch is an open-source deep-learning framework released by Meta in 2017, now governed by the PyTorch Foundation under the Linux Foundation. It is used both for research and for training and running large models in production.

Mask R-CNN is a 2017 framework for object instance segmentation. It detects objects and produces a high-quality mask for each instance by adding a parallel mask branch to Faster R-CNN.

A 2017 paper that introduced the Transformer, a neural network built entirely on attention with no recurrence. It set new machine translation results and became the architecture behind almost every large language model since.

AlphaGo Zero learned Go only by playing against itself, starting from random play, and beat the earlier champion-defeating AlphaGo 100 games to 0 after three days of training. The Nature paper appeared in October 2017.

A 2018 paper by Matthew E. Peters and five co-authors that represents each word with the internal states of a pre-trained deep bidirectional language model, so vectors change with context; adding them to existing models improved results on six NLP tasks.

OpenAI's 2018 paper introducing GPT, a Transformer decoder pre-trained to predict the next token and then fine-tuned for specific tasks. It improved the state of the art on 9 of 12 tasks studied and set the recipe the GPT series followed.

BERT is a language model from Google that reads text in both directions and is pre-trained by filling in masked words. Fine-tuned per task, it set new records on question answering and language-understanding benchmarks and later went into Google Search.

Google's 2018 BERT paper introduced an encoder-only Transformer that reads text in both directions and is pre-trained by hiding words. It set new results on eleven language tasks and made download-and-fine-tune the default workflow.

An NVIDIA paper from 2019 presenting an intra-layer model parallelism method, implemented with a few communication operations in native PyTorch, that trained transformer models of up to 8.3 billion parameters and improved several benchmarks.

A 2019 paper by Francois Chollet that defines intelligence as skill-acquisition efficiency and introduces the Abstraction and Reasoning Corpus (ARC), a benchmark of grid tasks that assumes no specialised knowledge or language.

A 2020 paper by Mildenhall, Srinivasan, Tancik, Barron, Ramamoorthi and Ng that represents a scene as a neural radiance field: a network mapping 5D coordinates to density and colour, optimized from a sparse set of images with known camera poses.

A 2020 paper that pairs a pre-trained seq2seq model with a dense Wikipedia index fetched by a neural retriever, setting state of the art on three open-domain QA tasks and producing more factual text than a parametric-only baseline.

GPT-3 is an autoregressive language model with 175 billion parameters, described in a May 2020 paper. It could perform new tasks from a few examples in the prompt, without any gradient updates, and it was first offered through OpenAI's API.

A benchmark released in 2020 that tests language models across 57 subjects, from elementary mathematics to law, and showed that even the largest models of the time were far from expert-level.

A 2020 paper showing that a plain Transformer applied directly to sequences of image patches, with no convolution at all, can match state-of-the-art convolutional networks on image classification.

A 2021 paper by Fedus, Zoph and Shazeer that simplifies Mixture-of-Experts routing so that language models can hold up to a trillion parameters while keeping per-example computation roughly constant.

A 2021 FAccT paper by Bender, Gebru, McMillan-Major and Shmitchell that asks whether language models can be too large and sets out recommendations on costs, data, research goals and alternative directions.

LoRA is a fine-tuning method that freezes a pre-trained model's weights and trains small low-rank matrices injected into each layer, cutting trainable parameters and memory by orders of magnitude.

A 2021 paper from OpenAI introduces Codex, a GPT model fine-tuned on public GitHub code, and HumanEval, an evaluation set that judges generated Python programs by whether they work rather than by how closely they match a reference.

RoseTTAFold is a three-track neural network for protein structure prediction, published by the Baker lab in 2021; its accuracy approached AlphaFold2's at CASP14, it models protein–protein complexes from sequence alone, and it is distributed as open code with a public web server.

The Nature paper described AlphaFold's methods, system design, and evaluation for protein-structure prediction.

A 2021 paper showing that finetuning a language model on many tasks described by instructions — instruction tuning — makes it much better at unseen tasks without examples, surpassing a larger model on most benchmarks tested.

A 2021 paper by Rombach, Blattmann, Lorenz, Esser and Ommer that runs diffusion models inside the latent space of pretrained autoencoders, cutting the compute needed for high-resolution image synthesis while adding cross-attention for text and bounding-box control.

InstructGPT is a family of GPT-3 models trained with reinforcement learning from human feedback so that they follow user instructions. OpenAI announced them in January 2022 and made them the default models in its API.

FlashAttention is an exact attention algorithm built around GPU memory traffic: by tiling computation so intermediates stay on chip, it cuts reads and writes to high bandwidth memory, training Transformers faster and enabling longer sequences.

ReAct is a prompting method that has a language model interleave written reasoning with calls to outside tools or environments, improving question answering, fact verification and interactive decision-making.

RT-1 is a transformer-based model for real-world robot control, trained on over 130,000 episodes collected with 13 robots and reported to perform more than 700 language instructions at a 97% success rate.

LLaMA is a family of foundation language models from 7B to 65B parameters, trained only on public datasets; its 13B model beat GPT-3 on most benchmarks, and all models were released to researchers.

OpenAI's technical report on GPT-4, submitted to arXiv in March 2023, describes a large multimodal Transformer model, reports human-level results on several professional benchmarks, and explains how the team made scaling behaviour predictable.

A 2023 paper that introduces the Segment Anything project: a promptable segmentation task, a model, and a dataset of over 1 billion masks on 11 million images, all released to advance foundation models for computer vision.

A crowdsourced benchmark in which anonymous pairs of language models battle and users vote, producing an Elo leaderboard; launched in 2023 and relaunched as LMArena in 2025.

A 2023 paper by Rafailov and colleagues that rewrites the reinforcement-learning-from-human-feedback objective so a language model can be aligned to preferences with a simple classification loss, without a separate reward model or reinforcement learning.

A July 2023 paper from Meta led by Hugo Touvron that introduces and releases Llama 2, a family of pretrained and fine-tuned language models from 7 to 70 billion parameters, together with a detailed account of how the chat versions were tuned and made safer.

A benchmark of 2,294 real software engineering problems taken from GitHub issues and the pull requests that resolved them, in which a model must edit a codebase to fix an issue; on release the best model solved under 2% of them.

Mamba is a sequence-model architecture introduced by Albert Gu and Tri Dao that replaces attention with selective state space models, giving linear-time scaling, faster inference, and language-model quality competitive with Transformers twice its size.

Mixtral 8x7B is a sparse mixture-of-experts language model from Mistral AI. A router sends each token to two of eight experts per layer, so the model holds 47B parameters but uses 13B during inference.

A report from Google's Gemini Team introducing the Gemini 1.5 family, multimodal models built around very long context, with Gemini 1.5 Pro recalling information across up to 10 million tokens.[^1]

Meta's 2024 paper describing Llama 3 as a family of foundation models — 8B, 70B and 405B parameters — trained on about 15 trillion tokens and reported to match leading models such as GPT-4 across a wide range of tasks.

A 2025 paper introducing ARC-AGI-2, a benchmark of input-output puzzle tasks built to measure abstract reasoning in frontier AI systems, with human testing as a baseline and a task set designed to give a more granular signal of progress.

Alignment is the problem of making a model's behaviour match what its users and designers actually intend, not just what its training objective literally rewards. Training on human preference comparisons is one practical route to it.

JAX is a Python library for accelerator-oriented array computation and program transformation, aimed at high-performance numerical computing and large-scale machine learning. It pairs a NumPy-style API with composable transformations for compilation, batching, differentiation and parallelization.

Reinforcement learning is the branch of machine learning and optimal control in which an agent learns, by trial and error, which actions in a dynamic environment maximise a reward signal.

Supervised learning trains a model to map inputs to outputs using paired labelled examples, so that it can predict answers for new, unseen data. Its central measure of success is how well the model generalises beyond the training set.

Unsupervised learning trains algorithms on data that carry no labels, letting them find structure such as clusters, low-dimensional representations or generative models. It underlies clustering tools and neural generative models from the Hopfield network and Boltzmann machine to the variational autoencoder.

ImageNet is a large, hierarchical image database built on the WordNet conceptual model and released at Princeton University in 2009. It holds more than 14 million images in 21,841 categories, with bounding-box annotations, and a 1,000-class subset used by the ILSVRC challenge.

Google's self-driving car prototype was a modified production car that used cameras, radar and a roof-mounted laser range finder to sense traffic, drove routes that had been mapped in advance by manually driven cars, and always carried a trained safety driver and a software operator.

AGI is the goal of AI systems that match or beat humans across a broad range of tasks. The term has no agreed definition; a framework by Morris et al. grades systems by performance level and generality.

Andrew Chi-Chih Yao is a computer scientist born in Shanghai who won the 2000 ACM A.M. Turing Award for fundamental contributions to the theory of computation, and later founded Tsinghua University's Yao Class and IIIS.

A vision-language-action model (VLA) is a robot policy that takes in camera images and a language instruction and outputs actions directly, usually by adapting a pretrained vision-language model to control.

Hallucination is false or misleading information that a model presents as fact. It is a central failure mode of text-generating systems and a main obstacle to trusting their output.

MYCIN was an early expert system built at Stanford in the 1970s to identify bacteria causing severe infections and recommend antibiotic treatment, notable for separating its rule base from its inference engine and for explaining its own reasoning.

A private Ivy League college in Hanover, New Hampshire, founded by royal charter in 1769. Its 1956 summer research project, proposed by John McCarthy, is credited with launching artificial intelligence as a field.

A public research university in Toronto where Geoffrey Hinton has been affiliated since 1987 and where AlexNet, the 2012 deep convolutional network, was developed.

A US public university founded in 1861, with campuses in Seattle, Bothell and Tacoma and a medical centre serving its state and region; it says it receives more federal research funding than any other US public university.

MIT, the Massachusetts Institute of Technology in Cambridge, Massachusetts, hosted the AI lab that Marvin Minsky and John McCarthy co-founded, which later became CSAIL. The lab also invented Lisp machines, and Minsky co-wrote the book Perceptrons.

The University of California, Berkeley is a public land-grant research university founded in 1868, the first land-grant university in California, known for a long Nobel record, a history of student activism and top rankings among public universities.

Stanford University is a private research university in California, home to the Stanford AI Lab (SAIL) and the Institute for Human-Centered AI, which publishes the AI Index.

Carnegie Mellon University in Pittsburgh has one of the earliest US computer science departments. Its researchers built the Logic Theorist, Sphinx speech recognition and the Deep Blue lineage, and won the 2007 DARPA Urban Challenge.

IBM is an American technology company founded in 1911 as the Computing-Tabulating-Recording Company and renamed International Business Machines in 1924. Its researchers built Deep Blue, which beat Garry Kasparov in 1997, and the Watson question-answering system.

Herbert A. Simon (1916–2001) was a political scientist, psychologist and computer scientist who, with Allen Newell and J. C. Shaw, built the first successful AI program and later received the Nobel Prize in Economics.

Berlin-born American computer scientist who wrote ELIZA, an early program that held a written conversation with a user, and later published an influential critique of artificial intelligence.

Bell Labs was the research arm of the Bell System, where the transistor, Unix and the C language were created. Yann LeCun's convolutional networks for handwritten digits were developed there, which made it an early home of neural-network research.

American computer scientist and psychologist who, with Herbert A. Simon and J. C. Shaw, built the first list-processing language and the earliest heuristic problem-solving programs, treating thought as computation.

Marvin Minsky (1927–2016) was an American scientist who co-founded MIT’s AI laboratory and co-signed the 1955 Dartmouth proposal. With Seymour Papert he wrote the 1969 book Perceptrons, which analysed the limits of simple neural networks.

Frank Rosenblatt (1928–1971) was an American psychologist who developed the perceptron, an early learning machine that adjusted its weights from examples. His Mark I Perceptron is now held by the Smithsonian Institution.

American physicist born in 1933 who invented the Hopfield network in 1982, a model that stores and recalls patterns using ideas from spin physics, and who received half of the 2024 Nobel Prize in Physics.

Edward Feigenbaum (born 1936) is known as the father of expert systems. His projects produced DENDRAL, META-DENDRAL, MYCIN and ONCOCIN, he pioneered the term "knowledge engineering", and he shared the 1994 ACM Turing Award.

Judea Pearl is a computer scientist and statistician who invented Bayesian networks and their main inference algorithms, and later built a mathematical framework for causal inference from data. He received the 2011 ACM Turing Award.

David Rumelhart (1942–2011) was an American cognitive psychologist. His 1986 work with colleagues helped bring back-propagation and neural-network models to a wide audience.

IBM Research built Deep Blue, which beat world chess champion Garry Kasparov in 1997, and Watson, which won Jeopardy! in 2011. Nathaniel Rochester of IBM was also one of the four signatories of the 1955 Dartmouth proposal.

Rodney Brooks (born 1954) is an Australian-born roboticist who promoted behaviour-based robotics and the subsumption architecture, directed MIT's AI Laboratory and CSAIL, and co-founded iRobot, Rethink Robotics and Robust.AI.

GPS was a program built at RAND to act as a universal problem-solving machine, and is credited as the first computer program to separate problem knowledge from a general solving strategy.

Fortran is a programming language for numerical and scientific computing, developed from 1954 and released by IBM in 1957. Its compiler showed that automatically translated code could run nearly as fast as hand-written machine code.

DARPA is the research agency of the United States Department of Defense, created in 1958 in response to Sputnik. It states its mission as creating and preventing technological surprise, and its Grand Challenges brought autonomous vehicle work to a wider community.

MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL), formed in 2003 by merging the MIT AI Lab with the Laboratory for Computer Science, is the largest laboratory on the MIT campus.

A checkers program built by Arthur Samuel at IBM, begun in 1949 and published in 1959, that learned by memorising positions, playing itself and scoring board positions, and played at a respectable amateur level.

Unimation was the Connecticut company founded in 1961 by George C. Devol and Joseph F. Engelberger that built the first Unimate robot and later the PUMA series.

Taiwan-born computer scientist and executive who built research labs at Microsoft and Google in China, co-founded Sinovation Ventures, wrote widely read books on AI, and founded the foundation-model company 01.AI.

Protein scientist at the University of Washington in Seattle, known for computational protein design: he designed a protein unlike any in nature in 2003 and received one half of the 2024 Nobel Prize in Chemistry.

The Association for Computational Linguistics (ACL) is the main scientific and professional society for natural language processing. Founded in 1962, it runs one of the field's two leading conferences and publishes the journal Computational Linguistics.

The Stanford Artificial Intelligence Laboratory, started by John McCarthy in 1963, is associated with the Stanford arm robot design, the Stanley self-driving car and the SAIL programming language.

German computer scientist and co-author, with Sepp Hochreiter, of the 1997 LSTM paper. He has directed the Swiss AI lab IDSIA since 1995 and has publicly argued that some contributions to AI deserve more credit.

A supervised maximum-margin model for classification and regression that separates classes by the widest possible margin and uses kernel functions to handle non-linear data.

ELIZA is an early program for holding a natural-language conversation with a computer, built at MIT by Joseph Weizenbaum in the mid-1960s. Its DOCTOR script imitated a psychotherapist, and the readiness of users to treat it as understanding became known as the ELIZA effect.

A January 1966 paper by Joseph Weizenbaum presenting ELIZA, a program on MIT's MAC time-sharing system that used keyword-triggered decomposition and reassembly rules to enable certain natural-language conversations between people and computers.

The Department of Machine Intelligence and Perception at the University of Edinburgh, created in October 1966 out of Donald Michie's Experimental Programming Unit, was an early British institutional home for artificial intelligence and a forerunner of the university's School of Informatics.

German computer scientist who co-developed the long short-term memory (LSTM) network, a recurrent neural network design for learning from sequences such as text and speech. He led the xLSTM work in 2024.

American computer scientist known for Google's core infrastructure — MapReduce, Bigtable and Spanner — and for leading work on Google Brain, TensorFlow and, from 2023, the role of Google's chief scientist.

IJCAI is a non-profit corporation founded in California in 1969 for scientific and educational purposes; it runs the IJCAI conference series (biennial in odd years from 1969, annual since 2015) and publishes a journal and proceedings.

PARC is a Silicon Valley research center founded by Xerox in 1970, known for Ethernet, laser printing and the Alto personal computer; it became an independent company in 2002 and joined SRI International in 2023.

The MIT Artificial Intelligence Laboratory, formed in 1970 under Marvin Minsky, grew out of MIT's earlier AI group inside Project MAC; in 2003 it merged with the Laboratory for Computer Science to form CSAIL.

SHRDLU was an early natural-language program built by Terry Winograd at MIT in 1968–1970 that conversed in English about a small blocks world, executing commands, answering questions and learning names — a landmark demo whose success fed an optimism later systems could not sustain.

Zhou Zhihua is a Chinese computer scientist at Nanjing University who works on ensemble learning, weakly supervised learning and multi-label learning. He founded the LAMDA institute there, serves as a university vice president, and was elected to the Chinese Academy of Sciences in 2025.

British computer scientist born in 1976 who specialises in reinforcement learning. He led DeepMind's AlphaGo project, co-authored the 2013 deep Q-network paper and later founded Ineffable Intelligence.

Stanford computer scientist Fei-Fei Li created ImageNet, the large-scale image dataset and challenge that gave data-driven computer vision a common benchmark, and has led AI research in both academia and industry.

Demis Hassabis is a British AI researcher and entrepreneur who co-founded DeepMind and was the last author of the 2016 AlphaGo paper. He shared the 2024 Nobel Prize in Chemistry for work on predicting protein structures.

R1, known inside Digital Equipment Corporation as XCON, was a rule-based expert system written by John P. McDermott at Carnegie Mellon University in 1978 to configure VAX computer orders automatically.

ICML is the annual International Conference on Machine Learning, held every year since 1980 and organized by the International Machine Learning Society. It is the oldest of the field's main conferences and one of its three most influential venues.

Czech computer scientist and lead author of word2vec, the 2013 methods that learn word vectors from text. His later work includes the fastText library and research on mathematical models that grow in complexity.

CVPR is the annual Conference on Computer Vision and Pattern Recognition, first held in 1983 and usually held in June in North America. It is known for selective, double-blind review and open-access papers.

A Boltzmann machine is a neural network whose configurations are scored by a global energy and trained by a stochastic learning algorithm. Introduced by Geoffrey Hinton and Terry Sejnowski, it brought energy-based models into cognitive science and machine learning.

Cyc is a hand-built knowledge base and ontology of common-sense knowledge, begun by Douglas Lenat in 1984 to give software the basic concepts and rules about how the world works.

American scientist (born 1985) who worked on AlphaFold at Google DeepMind and shared half of the 2024 Nobel Prize in Chemistry with Demis Hassabis for protein structure prediction.

A 1985 Cognitive Science paper by Ackley, Hinton and Sejnowski that presents a general learning rule for connectionist networks and reports simple examples of efficient internal representations.

A recurrent neural network (RNN) is a neural network with feedback connections that carry information from one step of a sequence to the next. Its long short-term memory variant (LSTM) was designed to keep useful memories over longer spans.

Brett Adcock is a founder of several technology companies, including the recruiting platform Vettery, the electric aircraft maker Archer Aviation and the humanoid robotics company Figure AI, where he is chief executive.

NeurIPS is an annual December conference on machine learning and computational neuroscience, founded in 1987 and counted among the most influential meetings in AI research.

NETtalk is an artificial neural network built by Terrence Sejnowski and Charles Rosenberg that learns to convert written English into phonemes and audible speech, showing that a network trained on examples can generalize to unseen words.

American AI researcher who co-founded OpenAI, led its reinforcement-learning team and helped develop the training methods behind ChatGPT; he later moved to Anthropic and then became chief scientist of Thinking Machines Lab.

A 3-layer back-propagation network that takes camera and laser range finder images and outputs the direction to steer, trained on simulated road images and tested on Carnegie Mellon's autonomous navigation test vehicle.

In 1989 LeCun and colleagues at AT&T Bell Laboratories trained a convolutional network with backpropagation to read handwritten zip-code digits from U.S. mail, reaching 5.0% test error. The work showed that the convolutional design could be learned end to end on a real task.

Dragon Dictate, released by Dragon Systems in 1990, was the first speech-recognition product sold to consumers; at $9,000 it was costly, but it started the line of products that led to continuous speech recognition.

Microsoft's corporate research division, founded in 1991, whose researchers produced work such as the ResNet image-recognition architecture and helped build features used in Kinect, Bing, HoloLens, Cortana and Microsoft Translator.

PC Therapist was Joseph Weintraub's conversational program; a modified version won the first annual Loebner Prize Competition in 1991, fooling five of 10 judges into thinking it was human.

A 1992 technical note by Watkins and Dayan that set out Q-learning, an incremental and computationally light way for an agent to learn optimal action values, and proved it converges to the optimum with probability 1.

Gerald Tesauro's 1992 paper tested whether temporal difference learning could master a complex real-world task, using self-play backgammon as the case study.

Chelsea Finn is a computer scientist and assistant professor at Stanford University whose research spans meta-learning, deep reinforcement learning and robot learning; she is a co-founder of Physical Intelligence.

Mila, the Quebec AI Institute, is a Montreal deep-learning institute founded in 1993 by Yoshua Bengio as LISA; its graduates include Ian Goodfellow and Hugo Larochelle.

C4.5 is a machine learning system that builds decision trees or if-then rules from cases with known classes. J. Ross Quinlan's 1993 book documents its C/UNIX implementation, source code, and use.

The Newton MessagePad was Apple's first handheld computer built on Newton technology, shipping on 2 August 1993 with a pen-based interface, handwriting recognition and a modular communications design.

AdaBoost, short for Adaptive Boosting, is a statistical classification meta-algorithm introduced by Yoav Freund and Robert Schapire in 1995. It reweights training examples, trains weak classifiers in sequence and combines their outputs into one weighted classifier.

Corinna Cortes and Vladimir Vapnik's 1995 paper presents a learning machine for two-group classification that maps inputs nonlinearly into a high-dimensional feature space, constructs a linear decision surface, and extends earlier work from separable to non-separable data.

Dragon NaturallySpeaking is speech recognition software from Dragon Systems, first released in 1997 as a continuous dictation product; it turns speech into text, runs spoken commands and reads documents aloud, and is still sold today under Microsoft.

NASA's Mars Pathfinder lander touched down on July 4, 1997, deployed the small Sojourner rover, used an airbag landing system for the first time, and returned thousands of images of the Martian surface.

A non-profit founded in 1998 that stewards the Open Source Definition and, since 2024, the Open Source AI Definition, setting a shared standard for what counts as open source.

MNIST is a database of 60,000 training and 10,000 test images of handwritten digits, normalized to 28×28 pixels and described in the 1998 paper by LeCun and colleagues as a shared benchmark for document recognition.

LeCun, Bottou, Bengio and Haffner’s 1998 review argued that learning systems should replace hand-designed features. It showed that convolutional networks such as LeNet-5 outperformed other methods on handwritten digits, and it introduced the MNIST dataset.

Brin and Page's PageRank technical report treats hyperlinks as citations to rank web pages and was used to build Google's search engine; it is one of the heavily cited papers on their early work.

Google's research division, based in Mountain View and usually dated to 2000, produced the Transformer, BERT and T5 — the work underlying modern large language models.

Leo Breiman's 2001 paper defines random forests as combinations of tree predictors whose random vectors are sampled independently from the same distribution, analyzes generalization error through tree strength and correlation, and reports error rates competitive with Adaboost and more robust to noise.

Torch is a scientific computing framework for machine learning built on the LuaJIT scripting language with a C/CUDA core, designed to put GPUs first and surrounded by a large ecosystem of community packages.

CALO was SRI's cognitive assistant project, begun in 2003 as part of a five-year, 150 million dollar DARPA program, built to learn from observation rather than fixed programming. Its work on virtual personal assistants was later leveraged to create Siri.

Topic models represent documents as mixtures of latent topics, giving a compact statistical description of text collections. The 2003 Latent Dirichlet Allocation paper formalized this idea as a hierarchical Bayesian model.

Published in 2003 by Yoshua Bengio and colleagues, this paper proposed learning a vector for every word together with a neural network that predicts the next word. It introduced the core idea behind word embeddings and reached lower test perplexity than n-gram models.

A word embedding represents each word as a list of numbers learned from text, so that words used in similar ways get similar lists. Embeddings are a basic building block of modern language models.

An independent, non-profit research institute in Cambridge, Massachusetts, founded in 2004 with support from Eli and Edythe Broad, working to understand the roots of disease and turn biological insights into impact for patients.

BigDog is a four-legged robot built by Boston Dynamics with DARPA funding, designed as a robotic pack mule that could carry loads over terrain too rough for conventional vehicles.

Jeffrey Dean and Sanjay Ghemawat's 2004 OSDI paper presents MapReduce, a programming model in which users write a map function and a reduce function while the run-time system parallelizes the work across large clusters of commodity machines.

OWL is a W3C Recommendation published on 10 February 2004: a semantic markup language for publishing and sharing ontologies on the Web, giving software formal definitions of terms so it can process the meaning of information, not just display it.

Connectionist temporal classification (CTC) is a method for training recurrent neural networks to label unsegmented sequences, such as speech or handwriting, without frame-by-frame alignment. Introduced in 2006, it let CTC-trained LSTMs win the first international contests for recurrent networks.

Theano is a Python library and optimizing compiler for mathematical expressions, especially matrix-valued ones, developed at the Montreal Institute for Learning Algorithms. Released in 2007, it combined symbolic differentiation, GPU execution and C code generation for numerically heavy research and teaching.

Common Crawl is a nonprofit that has archived the public web since 2007 and gives its petabyte-scale datasets away for free. Its corpus became a core source of training data for large language models.

CIFAR-10 and CIFAR-100 are labeled image datasets drawn from the 80 million tiny images collection, documented in a 2009 technical report by Alex Krizhevsky, Vinod Nair and Geoffrey Hinton. CIFAR-10 holds 60,000 32×32 colour images in 10 classes.

A 2009 technical report by Alex Krizhevsky that introduces CIFAR-10 and CIFAR-100 and studies multi-layer generative models for learning features from tiny colour images.

Deep neural network acoustic modeling uses multilayer networks to map speech to units such as phones. A 2010 NIPS paper used Deep Belief Nets and a mean-covariance RBM to extract speech features, reaching 20.5% phone error on TIMIT.

An open-source Python library for statistical learning, begun in 2007, first released publicly on 1 February 2010 by a team at Inria, and now one of the most popular machine learning programs on GitHub.

Graphics processing units (GPUs) run many arithmetic operations in parallel, which suits neural-network training. Their use from 2010 onward made deep learning practical at much larger scales.

Xavier (Glorot) initialization is a weight-initialization scheme from Glorot and Bengio's 2010 paper that sets each layer's starting weight scale so signals stay stable across layers, giving substantially faster convergence in deep feedforward networks.

A consumer graphics card NVIDIA released in November 2010 on its Fermi architecture. Two of them, each with 3 GB of memory, later trained AlexNet, the network that won ILSVRC-2012 and which NVIDIA's timeline credits with sparking the era of modern AI.

A 2010 NIPS paper by Dahl, Ranzato, Mohamed and Hinton applies Deep Belief Nets and a mean-covariance RBM to acoustic modeling, reaching 20.5% phone error on speaker-independent TIMIT phone recognition.

A 2011 workshop paper by Ronan Collobert, Koray Kavukcuoglu and Clement Farabet describing Torch7, a Lua-based numeric computing framework and machine learning library built for flexible experimentation on fast low-level kernels.

Google Brain was a deep-learning research team started inside Google's X in 2011. Its cat-recognition experiment, TensorFlow, sequence-to-sequence models and the Transformer made it central to modern AI; it merged into Google DeepMind in 2023.

A 2011 Journal of Machine Learning Research paper by Collobert, Weston, Bottou, Karlen, Kavukcuoglu and Kuksa, presenting one neural network architecture and learning algorithm for part-of-speech tagging, chunking, named entity recognition and semantic role labeling, trained on mostly unlabeled data with little task-specific engineering.

Robonaut 2 is a humanoid robot built jointly by NASA and General Motors that flew to the International Space Station in 2011 to show how dexterous robots behave in orbit, and later gained legs to move around inside the station.

A free beta software development kit released by Microsoft on 16 June 2011 that let developers use the Kinect sensor's depth sensing, motion tracking and voice recognition on Windows PCs, initially for non-commercial work.

AlexNet is the eight-layer convolutional network that won the ILSVRC2012 image classification task by a wide margin, showing that a large network trained on a large labelled dataset with GPUs could beat hand-designed features.

ICLR is an annual machine learning conference held in late April or early May, founded in 2012 by Yann LeCun and Yoshua Bengio, and known for its open peer review process.

A research institute at the University of Washington, founded in 2012 and led by David Baker, that designs new proteins for medicine and technology; its labs produced RoseTTAFold and, it says, the first computationally designed vaccine.

In 2012, Quoc Le, Andrew Ng and colleagues trained a large unsupervised neural network on 10 million internet images and showed it could detect faces without a single labelled face.

A 2012 article in IEEE Signal Processing Magazine in which four research groups argued that deep neural networks, replacing Gaussian mixture models, gave better acoustic models for speech recognition.

A variational autoencoder is a generative model whose encoder maps an input to a distribution over a latent space and whose decoder reconstructs the input from a sample; training maximises a lower bound on the data likelihood.

Submitted to arXiv in January 2013 by Tomas Mikolov and colleagues at Google, this paper introduced CBOW and Skip-gram, two simple architectures that learn word vectors quickly from very large text collections. They are the basis of the word2vec project.

A 2013 paper by Mikolov, Sutskever, Chen, Corrado and Dean presents continuous Skip-gram as an efficient method for learning distributed word and phrase representations, with extensions that improve vector quality and training speed.

R-CNN, posted to arXiv by Girshick, Donahue, Darrell and Malik in November 2013, applied convolutional neural networks to region proposals for object detection and lifted mean average precision on PASCAL VOC 2012 to 53.3%, more than 30% above the previous best[^1].

Zeiler and Fergus ask why large convolutional networks perform so well on ImageNet, introduce a visualization technique for intermediate layers, and use ablation to find an architecture that outperforms Krizhevsky et al. The model also transfers to Caltech-101 and Caltech-256.

A 2013 paper by Mnih, Kavukcuoglu, Silver and colleagues showed a convolutional neural network trained with a Q-learning variant learning to play seven Atari 2600 games directly from raw pixels, beating prior methods on six and human experts on three.

A DeepMind reinforcement-learning agent that combined Q-learning with a convolutional neural network to play Atari 2600 games straight from raw pixels, reaching human-level performance on many titles without being told the rules.

A 2013 paper by Diederik P. Kingma and Max Welling that introduced a stochastic variational inference and learning algorithm for directed latent-variable models with intractable posteriors, making approximate inference practical on large datasets.

DeepFace is a 2014 face verification pipeline that rebuilt the alignment and representation stages of the standard face pipeline, reaching 97.35% accuracy on Labeled Faces in the Wild with a nine-layer network of over 120 million parameters.

A 2014 paper by Srivastava, Hinton, Krizhevsky, Sutskever and Salakhutdinov introducing dropout, a training trick that randomly removes units so large neural networks overfit less.

Nonprofit AI research institute founded by Paul Allen in 2014 and based in Seattle, known for Semantic Scholar and for the openly released OLMo family of language models, whose training data, code and checkpoints are published alongside the weights.

A 2014 paper introducing Microsoft COCO, a large dataset of everyday scenes in which common objects are labeled with per-instance segmentations, built to study object recognition as part of scene understanding.

A 2014 paper from Google researchers that introduced the Inception architecture, whose GoogLeNet model set a new state of the art in ILSVRC 2014 by widening and deepening a network within a fixed computational budget.

A 2014 paper by Graves, Wayne and Danihelka that couples a neural network to an external memory it reads and writes through attention, making the whole machine differentiable and trainable by gradient descent.

A 2014 paper presenting an end-to-end deep-learning speech recognition system that removes hand-engineered noise modules and the phoneme dictionary, relying instead on recurrent networks trained across multiple GPUs and synthesized data.

A technique that re-centres each layer's inputs to zero and rescales them to unit variance over a mini-batch, making deep networks faster and more stable to train.

A 2015 paper by Sergey Ioffe and Christian Szegedy that proposed normalising each layer's inputs per training mini-batch, which allowed higher learning rates, looser initialisation and much faster training of deep networks.

A convolutional network for biomedical image segmentation, posted to arXiv on May 18, 2015, that trains end to end from very few annotated images and won the ISBI cell tracking challenge 2015 in its categories.

A 2015 paper that introduced a Region Proposal Network sharing full-image convolutional features with the detection network, making proposals nearly cost-free and enabling end-to-end training, 5 fps GPU detection, and state-of-the-art accuracy.

The 2015 YOLO paper recast object detection as a single neural network's regression over bounding boxes and class probabilities, achieving real-time detection while reducing false positives on empty background.

The 2015 paper introduced YOLO, which treats object detection as a single regression from image pixels to bounding boxes and class probabilities. One neural network makes predictions in one evaluation, reaching real-time speed.

The 2015 paper introduced residual networks, which add each layer's input back to its output through a shortcut connection. This made networks of over a hundred layers trainable and won first place in several ILSVRC and COCO tasks in 2015.

Boston Dynamics' 2016 generation of the Atlas humanoid, called Atlas HD or hydraulic Atlas: smaller, lighter and more agile than earlier versions, electrically powered and hydraulically actuated, and retired in 2024.

The 2016 Nature paper describing AlphaGo, which combined deep neural networks with tree search. In October 2015 its distributed version beat the European Go champion Fan Hui 5-0, the first time a program defeated a professional player in full-sized Go without handicap.

A 2016 analysis of residual networks arguing that with identity skip connections and the activation placed after addition, signals travel directly between blocks; a new residual unit built on this eases training, and a 1001-layer network reaches 4.62% error on CIFAR-10.

The Tesla P100, announced by NVIDIA in April 2016, was the first data-center GPU on the Pascal architecture, and the first to combine stacked HBM2 memory, half-precision math aimed at deep learning and the NVLink interconnect.

OpenAI Gym, announced as a beta in April 2016, is a toolkit for developing and comparing reinforcement learning algorithms: a growing suite of environments, from classic control tasks to Atari games and simulated robots, plus a website for comparing and reproducing results.

SQuAD is a reading-comprehension dataset of more than 100,000 crowdworker questions on Wikipedia articles, each answered by a span of its passage. Released in 2016, it gave machine comprehension a large, freely available benchmark with a measured human ceiling.

A DeepMind generative model that predicts raw audio one sample at a time, bringing neural text-to-speech close to human quality and later powering Google Assistant and Cloud Text-to-Speech voices.

A 2016 paper describing Google's neural machine translation system, GNMT: a deep LSTM encoder-decoder with attention, designed to make end-to-end translation practical by addressing rare words, parallelism and inference cost.

Deepfake is synthetic media: images, video or audio made or altered with AI. The term, a blend of "deep learning" and "fake", appeared in late 2017 and now names a whole class of face-swapping and manipulation techniques.

A 2017 paper showing that an AI agent can learn a task from short human comparisons instead of a hand-written reward. The method, reinforcement learning from human feedback, later became central to training chat assistants.

A 2017 paper by Schulman, Wolski, Dhariwal, Radford and Klimov that introduced proximal policy optimization (PPO), a reinforcement learning method that is much simpler to implement and tune and became OpenAI's default RL algorithm.

A 2017 paper by Lin, Goyal, Girshick, He and Dollar that argues dense one-stage detectors are held back by extreme foreground-background imbalance and proposes focal loss, a reshaped cross-entropy that down-weights well-classified examples.

A DeepMind Go program that learned the game from scratch by playing against itself, using no human game records, and beat earlier versions of AlphaGo within days.

mixup is a training principle from a 2017 paper by Zhang, Cisse, Dauphin and Lopez-Paz that trains networks on convex combinations of pairs of examples and their labels, improving generalisation on image, speech and tabular benchmarks.

A 2017 paper by van den Oord, Vinyals and Kavukcuoglu proposing VQ-VAE, a generative model whose latent code is discrete rather than continuous, and whose prior is learned rather than fixed, so that useful representations can be learned without supervision.

A 2017 arXiv paper, later published in Science, reporting that a single self-play reinforcement learning algorithm, AlphaZero, mastered chess, shogi and Go and beat strong specialised programs in each.

AlphaZero is a general game-playing system from DeepMind, described in a preprint in December 2017, that reached a superhuman level in chess, shogi and Go within 24 hours starting from random play, using only the rules of each game.

An annual artificial intelligence conference held in Shanghai since 2018, co-hosted by Chinese government bodies and the Shanghai municipal government, combining forums, exhibitions and signed investment projects.

ULMFiT is a 2018 transfer-learning method by Jeremy Howard and Sebastian Ruder for fine-tuning language models across NLP tasks; it beat the state of the art on six text-classification tasks and performed well with few labels.

GLUE is a benchmark and analysis platform that evaluates natural language understanding models across a diverse set of existing tasks, with an online comparison platform and a hand-crafted diagnostic suite for linguistic analysis.

OpenAI Five was a team of five reinforcement-learning bots that played the video game Dota 2; in April 2019 it beat OG, the reigning world champions, in a best-of-three series.

GPT-1 was OpenAI's first generative pre-trained transformer, published in June 2018. It showed that one language model, pre-trained on unlabelled text and then fine-tuned, could improve results on many language tasks.

A 2018 paper by Chen, Rubanova, Bettencourt and Duvenaud that replaces a fixed sequence of hidden layers with a neural network describing the derivative of the hidden state, so outputs are computed by an ODE solver — yielding continuous-depth models with constant memory cost.[^1]

A free, open catalogue that links machine learning papers to their code and maintains benchmark leaderboards tracking the state of the art; built by Atlas ML, majority-owned by Facebook from December 2019, and relaunched as paperswithcode.co by Hugging Face in 2026.

DeepMind's first protein-structure prediction system, which placed first overall in the 2018 CASP13 blind assessment by estimating probability distributions of distances between residues.

AlphaStar is a DeepMind system that plays StarCraft II. In 2019 it reached Grandmaster level with all three races, ranking above 99.8% of active players on Battle.net.

Baidu's first ERNIE pre-trained language model, released in March 2019, added entity- and phrase-level masking to BERT-style pre-training and reported state-of-the-art results on five Chinese NLP tasks.

SuperGLUE is a benchmark for general-purpose language understanding, released in 2019 as a harder successor to GLUE. It bundles eight tasks, a single-number metric, an analysis toolkit and a public leaderboard.

XLNet is a generalized autoregressive pretraining method from a 2019 paper. It learns bidirectional contexts by maximizing expected likelihood over all factorization-order permutations and outperforms BERT on 20 tasks under comparable settings.

A replication study of BERT pretraining that measures how much key hyperparameters and training data size matter, finds BERT was significantly undertrained, and shows a properly trained BERT matches or beats later models, reaching state of the art on GLUE, RACE and SQuAD.

Model parallelism splits one model's parameters and computation across multiple accelerators, so networks too large for a single device can still be trained. NVIDIA's 2019 Megatron-LM paper showed an intra-layer version built in native PyTorch.

A 2019 paper that casts every language task as text-to-text, systematically compares transfer-learning choices, releases the C4 corpus and models, and reports state-of-the-art results on summarization, question answering, and classification.

MuZero is a DeepMind reinforcement-learning algorithm that learns a model of its environment and plans inside that model, mastering Atari, Go, chess and shogi without being given the rules of those games.

A 2020 paper by Clark, Luong, Le and Manning that pre-trains text encoders by detecting replaced tokens instead of filling in masked ones, matching much larger baselines with far less compute.

GShard, submitted to arXiv in June 2020, is a set of annotation APIs and an XLA compiler extension from Google that lets existing model code express parallel computation, scaling a sparsely gated mixture-of-experts translation model past 600 billion parameters.

EleutherAI is a grassroots AI research collective that began as a Discord server in July 2020. It released The Pile and the open GPT-style models GPT-Neo, GPT-J and GPT-NeoX-20B, and in 2023 it became a nonprofit research institute.

A 2020 paper by Alexey Dosovitskiy and colleagues showing that a pure Transformer applied directly to sequences of image patches — the Vision Transformer, or ViT — matches state-of-the-art convolutional networks on image classification when pre-trained on large amounts of data.

AlphaFold2 is DeepMind's protein structure prediction system, announced in 2020. It reached accuracy competitive with experimental methods in many cases, and work in this area received half of the 2024 Nobel Prize in Chemistry.

The Pile is an 825 GiB English text dataset of 22 combined sources, released by EleutherAI on 31 December 2020 with an arXiv paper. It became the training data for several of the group's open models.

A state space model represents a sequence as a hidden state that evolves with the input. S4 brought the idea into deep learning in 2021, and Mamba made its parameters input-dependent, enabling linear-time modeling of long sequences.

A 2021 OpenAI paper describing a single 12-billion-parameter autoregressive transformer trained on 250 million image-text pairs, which generates images from text and, evaluated zero-shot, is competitive with domain-specific models.

MATH is a dataset of 12,500 challenging competition mathematics problems, each with a full step-by-step solution. Released in 2021, it was built to measure and train mathematical reasoning in models.

LaMDA is a family of conversational language models that Google announced in 2021 and described in a 2022 paper, notable for its work on safety and factual grounding in dialogue and for later powering Bard.

GPT-J-6B is a six-billion-parameter open language model released by EleutherAI in June 2021 under the Apache 2.0 license. Its published scores sit close to GPT-3 Curie, making it a widely used free baseline for large-model research.

ERNIE 3.0 is a 10-billion-parameter pre-trained language model from 2021 that trains on plain text plus a large-scale knowledge graph and fuses auto-encoding and auto-regressive networks, so one model handles both language understanding and generation.

OpenAI Codex is a 2021 GPT language model fine-tuned on public code from GitHub. It translates natural language into code, and was introduced alongside the HumanEval benchmark for testing code correctness.

A 2021 arXiv paper by Austin, Odena, Nye and eight other authors that introduced MBPP, a benchmark of 974 short Python programming tasks, and measured how language models from 244M to 137B parameters turn natural-language descriptions into working code.

GSM8K is an OpenAI benchmark of 8.5K grade school math word problems, released with a paper showing that even the largest Transformer models scored poorly, and that training verifiers to select among many candidate solutions improved results.

Adept is an AI company founded in 2022 that set out to train large multimodal models and turn them into agents, open-sourced the Fuyu-8B model in 2023, and shifted focus in 2024 when its co-founders joined Amazon.

GPT-NeoX-20B is a 20-billion-parameter open autoregressive language model released by EleutherAI in February 2022, with weights under Apache 2.0 and its training code, described by its makers as the largest publicly accessible general-purpose model of its kind at the time.

PaLM is a 540-billion-parameter dense Transformer language model announced by Google Research in April 2022, trained with the Pathways system across 6,144 TPU v4 chips; it set strong few-shot results on many benchmarks.

Imagen is a text-to-image diffusion model that uses a large pretrained language model to encode prompts; it reached a zero-shot FID of 7.27 on COCO and introduced the DrawBench benchmark.

Imagen is a text-to-image diffusion model announced by Google Research in May 2022, known for pairing a frozen large language model with diffusion and for the DrawBench benchmark; Google did not release it publicly.

BIG-bench is a large community-built benchmark of 204 tasks designed to test language models on problems believed to lie beyond their current abilities, described in a paper posted in June 2022.

A 2022 paper arguing that some abilities of large language models appear only once models are large enough, and cannot be predicted by extrapolating from smaller models — so further scaling may keep widening what these models can do.

BLOOM is a 176-billion-parameter open multilingual language model released on 12 July 2022 by the BigScience project, with weights available under a responsible-use license.

GLM-130B is a 130-billion-parameter open bilingual Chinese–English dense language model from Tsinghua University's KEG group and Zhipu AI, built to give researchers an open model that matches GPT-3 at the hundred-billion scale.

A 2022 paper by Percy Liang, Rishi Bommasani, Tony Lee and many co-authors that ran 30 language models through the same 42 scenarios and seven metrics, turning scattered benchmarks into one standardized evaluation.

Constitutional AI, a 2022 paper from Anthropic, proposes training a harmless AI assistant with AI-generated feedback guided by a written set of principles, instead of human labels of harmful outputs.

A 2022 paper by William Peebles and Saining Xie that replaces the U-Net backbone of diffusion models with a transformer operating on latent patches, showing that image quality improves predictably as compute grows and reaching a state-of-the-art FID of 2.27 on ImageNet 256×256.

A public Hugging Face leaderboard that ran a fixed set of benchmarks on open language models in one identical setup, from 2023 to 2025, so their scores could be reproduced and compared.

A 2023 paper by Schick, Dwivedi-Yu, Dessì and co-authors showing that a language model can teach itself when to call a calculator, search engine, Q&A system, translation system or calendar, and how to use the results.

A March 2023 paper by Microsoft researchers testing an early GPT-4, arguing it shows strikingly general, near-human performance across many domains and could be seen as an early but incomplete step toward AGI.

QLoRA is a finetuning method that trains a 65B-parameter language model on a single 48GB GPU while matching full 16-bit finetuning quality, using 4-bit quantization, double quantization and paged optimizers.

Sakana AI is a Tokyo research and development company founded in July 2023 by David Ha, Ren Ito and Llion Jones, known for evolutionary model merging, projects such as The AI Scientist, and Japanese-language models named Namazu.

FlashAttention-2 is a 2023 revision of the FlashAttention attention kernel that reorganizes work across GPU thread blocks and warps, roughly doubling speed and reaching up to 73% of the A100's theoretical peak throughput.

WebArena is a reproducible environment of four fully functional websites — e-commerce, a social forum, collaborative software development and content management — built for testing language-guided web agents. Its best GPT-4 agent completed 14.41% of tasks; humans completed 78.24%.[^1]

GPQA is a benchmark of 448 expert-written multiple-choice questions in biology, physics and chemistry, built so that skilled non-experts cannot look the answers up, and hard enough that the strongest GPT-4 baseline scored 39% while domain experts reached 65%.

GAIA is a benchmark of real-world questions that are easy for humans but hard for AI assistants, testing reasoning, multimodality, web browsing, and tool use. It compares AI progress against average human performance.

OLMo is the Allen Institute for AI's open language model, released on 1 February 2024 in 1B and 7B sizes with its Dolma training data, training code, logs and checkpoints, all under Apache 2.0.

A benchmark for code LLMs that keeps collecting fresh contest problems from LeetCode, AtCoder and CodeForces, so that scores reflect ability rather than memorised test data, and that also tests code repair and execution reasoning.

EvoLLM-JP is a Japanese model from Sakana AI built by evolutionarily searching merges of three open-source models instead of training from scratch; the 7B result has mathematics ability none of its sources had alone.

Jamba, released by AI21 Labs on 28 March 2024, is an open-weights model that combines Mamba state-space layers with Transformer attention. It has 52 billion total parameters and a 256k-token context window.

OSWorld is a benchmark and environment for testing multimodal agents on 369 real computer tasks in Ubuntu, Windows and macOS; humans complete over 72.36% of the tasks, while the best model reaches only 12.24%.

Tri Dao and Albert Gu's paper introduces state space duality, a framework showing that structured state space models and attention variants are closely related, and uses it to derive Mamba-2, whose core layer runs 2 to 8 times faster than Mamba's.

MMLU-Pro is a 2024 benchmark that makes the widely used MMLU knowledge test harder: its questions are reasoning-focused and each has ten answer options instead of four.

τ-bench is a benchmark, submitted to arXiv in June 2024, that tests language agents in dynamic conversations with a simulated user, with domain-specific API tools and policy rules, and scores them by database state rather than by the text of their replies.

A public benefit AI company that introduced itself alongside its ESM3 model in June 2024 and whose team joined Biohub in November 2025.

FlashAttention-3 is an attention algorithm for NVIDIA Hopper (H100) GPUs that reaches up to 75% utilization in FP16 and close to 1.2 PFLOPs/s in FP8, using asynchronous execution and low-precision techniques.

SAM 2 is a Meta FAIR foundation model that segments objects in images and videos from simple prompts such as points, boxes or masks, released together with a large video segmentation dataset and training code.

LMSYS (Large Model Systems) is a non-profit that started as a multi-university collaboration in 2023 and became an incorporated organisation in September 2024, building open models, serving systems and evaluation platforms such as Chatbot Arena, SGLang, FastChat and Vicuna.

FrontierMath is a benchmark of hundreds of original, unpublished mathematics problems written and vetted by expert mathematicians, introduced by Epoch AI to test advanced mathematical reasoning in AI. At release, state-of-the-art models solved under 2% of the problems.[^1][^2]

Humanity's Last Exam (HLE) is a 2,500-question, multi-modal benchmark spanning dozens of academic subjects, built by experts worldwide so that frontier language models can still be measured after saturating older tests.

A paper showing that a 32B model fine-tuned on 1,000 carefully chosen questions and run with a simple "budget forcing" trick can beat o1-preview on competition maths by up to 27%, with model, data and code released.

A research institute of Abu Dhabi's Mohamed bin Zayed University of Artificial Intelligence, launched in May 2025, working from Abu Dhabi, Silicon Valley and Paris on open frontier foundation models including K2, Jais and PAN.

A 2025 paper proposing Absolute Zero, a reinforcement-learning paradigm in which one model writes its own code-reasoning tasks and then solves them, with a code executor as the only judge, reaching state of the art on coding and maths without external training data.

Terminal-Bench 2.0 is a curated benchmark of 89 hard tasks carried out in computer terminal environments, built to measure how well AI agents handle realistic command-line work; frontier models and agents score below 65% on it.

Honda Motor Co., Ltd. is the company that built ASIMO, the humanoid robot it announced in Tokyo on 20 November 2000, following humanoid robotics research that began in 1986 and ran through a series of experimental walking machines.

SRI International is a non-profit research institute in Menlo Park, California, founded in 1946 as the Stanford Research Institute. Its AI Center built Shakey the robot, and its later work led to Siri.

Chris Olah is a Canadian machine learning researcher and co-founder of Anthropic, known for work on mechanistic interpretability, the effort to understand the internal workings of neural networks.

A loss function turns the cost of a prediction into a single number, and training a model means making that number as small as possible. Its choice decides which mistakes a model is punished for.

Overfitting is when a model fits its training data so closely that it performs poorly on new data. It is one half of the bias–variance trade-off and is usually spotted by comparing training and validation performance.

Regularization is the practice of making a problem's answer simpler, usually by adding a penalty term to the objective, so that ill-posed problems become solvable and models avoid overfitting.

A function applied to each node's weighted sum, giving neural networks their non-linearity. Sigmoid, tanh and ReLU shaped early deep learning; GELU now dominates transformers.

Embodied AI is the study of agents that learn by interacting with an environment, using egocentric perception, rather than only from datasets curated from the internet. Simulators serve as virtual testbeds where such agents are trained and evaluated.

Benchmark contamination means the test items a model is evaluated on, or close variants of them, were present in its pretraining data, so scores partly measure memory of seen text rather than the ability the benchmark is meant to probe.

Mechanistic interpretability explains model behaviour by finding the internal computations and features that produce it. Work on superposition and sparse autoencoders shows how many features can be packed into few neurons.

Reward hacking is what happens when a system optimises a proxy reward that only imperfectly captures what we want: it scores well on the proxy while performing badly on the true objective.

The RoboCup Federation runs an international robot-soccer initiative and competition whose stated aim is to develop robotic players able to beat a human World Cup champion team.

Baidu Research is the research arm of Baidu, built around the Institute of Deep Learning founded in 2013, with laboratories in Beijing and Silicon Valley; in 2014 Andrew Ng became chief scientist in charge of the whole organisation.

Jan Leike is an AI alignment researcher who joined Anthropic in May 2024 after leaving OpenAI, where he had co-led the superalignment project with Ilya Sutskever.

Uber ATG was Uber's self-driving unit. After a fatal 2018 crash with a pedestrian that the NTSB traced to a distracted safety operator and weak safety oversight, Uber sold the group to Aurora in December 2020.

The Semantic Web is a W3C standards effort that uses RDF and OWL to describe data so different applications, enterprises and communities can share and reuse information, moving from research toward practical tools.

AutoNav is the autonomous driving software used by NASA's Spirit and Opportunity Mars rovers. It combines stereo vision, GESTALT terrain assessment, Visual Odometry, and Field D* path planning to let rovers choose and drive safe paths without step-by-step direction from Earth.

Visual odometry lets a rover measure how far it actually moved by tracking features in camera images taken before and after a short drive, instead of trusting wheel turns. NASA's Mars Exploration Rovers were the first to demonstrate it.

A four-wheeled planetary rover prototype built at Carnegie Mellon's Robotics Institute with NASA funding, Nomad tested autonomous driving and rock analysis in Antarctica, hunting meteorites on the ice without step-by-step human control.

PackBot is iRobot's tactical mobile robot, developed under a 1998 DARPA contract. It carried out search work at the World Trade Center site in 2001 and was first deployed by the US military in 2002.

Chinook is a checkers-playing computer program built by Jonathan Schaeffer and colleagues [^1]. It became the first computer program to win a human world championship in any game, and its completed form cannot be beaten [^1].

Jared Kaplan is a physicist turned AI researcher who co-founded Anthropic and serves as its Chief Science Officer. He was first author of the 2020 paper “Scaling Laws for Neural Language Models.”

Alec Radford is a machine learning researcher who worked at OpenAI from 2016 to 2024. He was first author of the GPT-2, CLIP and Whisper papers, and lead author of the 2018 paper that introduced the GPT approach.

Noam Shazeer is an AI researcher who co-wrote the Transformer paper and was first author of the paper on sparsely gated mixture-of-experts layers. He co-founded Character.AI, returned to Google in 2024, and announced a move to OpenAI in 2026.

Ashish Vaswani is a computer scientist best known as the first-listed author of “Attention Is All You Need,” the 2017 paper that introduced the Transformer architecture. He also co-founded the AI companies Adept and Essential AI.

Kaiming He is a computer-vision and deep-learning researcher, first author of the 2015 ResNet paper that introduced residual connections, now a tenured associate professor at MIT.

Aidan Gomez is a Canadian computer scientist, co-founder and CEO of Cohere, and one of the eight authors of the 2017 paper that introduced the Transformer architecture.

Andrew G. Barto is a computer scientist who helped establish reinforcement learning as a general framework and shared the 2024 ACM A.M. Turing Award with Richard Sutton.

Richard Sutton is a computer scientist at the University of Alberta known for his work on reinforcement learning; he co-authored the field's standard textbook and shared the 2024 ACM A.M. Turing Award with Andrew Barto.

Raj Reddy is an Indian-born computer scientist who built the first system capable of continuous speech recognition and led major computing institutions at Carnegie Mellon University. He shared the 1994 ACM Turing Award with Ed Feigenbaum.

Epoch AI's Benchmarking Hub collects benchmark results for leading AI models, including Epoch-run evaluations, and combines many scores into the Epoch Capabilities Index, a single general capability scale.

Sergey Levine is a computer scientist who studies how robots can learn control behaviour from data. He is a professor at UC Berkeley and a co-founder of Physical Intelligence.

The University of Alberta's Department of Computing Science is the institutional home of the Chinook checkers project, and the university has trained a long line of researchers who use games to test ideas in artificial intelligence.

American electrical engineer who built one of the first checkers-playing programs and, in a 1959 paper, helped name the field now called machine learning.

Koray Kavukcuoglu is a DeepMind computer scientist who co-authored the first deep Q-network paper and later, as Google DeepMind's CTO and Google's chief AI architect, fronted successive Gemini model launches.

Durk Kingma is a machine learning researcher and co-author of the variational autoencoder and the Adam optimizer; he has worked at OpenAI, Google Brain / DeepMind and Anthropic.

Ian Goodfellow is a machine learning researcher and first author of the 2014 paper that introduced generative adversarial networks, which train a generator and a discriminator against each other. He has worked at OpenAI, Google Brain, Apple and Google DeepMind.

Quoc Le is a machine-learning researcher whose 2013 Stanford Ph.D. thesis was on scalable feature learning. He co-authored papers on large-scale unsupervised learning, sequence-to-sequence models, neural architecture search and EfficientNet; the sequence-to-sequence paper won a NeurIPS 2024 Test of Time award.

Niki Parmar is one of the eight authors of "Attention Is All You Need", the 2017 paper that introduced the Transformer architecture, and is named in Google's own account of that work.

Llion Jones is one of the eight authors of the 2017 paper "Attention Is All You Need" and a co-founder and CTO of Sakana AI.

Illia Polosukhin is one of the eight authors of the 2017 Transformer paper and a co-founder of NEAR Protocol. A 2024 NEAR post identifies him as CEO of the NEAR Foundation.[^1][^3][^6]

Harry Shum is a computer scientist who spent most of his career at Microsoft, helping found Microsoft Research Asia in Beijing and later leading the company's AI and Research group. He is an IEEE Fellow and ACM Fellow recognised for computer vision and graphics.

Wang Jian is the founder of Alibaba Cloud and a member of the Chinese Academy of Engineering. He was previously a Zhejiang University psychology professor and Microsoft Research Asia executive, and directs Zhijiang Laboratory.

Stanford computer science professor who directs the Center for Research on Foundation Models and works on making language models more open and more rigorously benchmarked.

Christopher Manning is a Stanford professor of linguistics and computer science, former SAIL director, and former ACL president, recognized with fellowships and academy memberships.

Pieter Abbeel is a UC Berkeley professor who directs the Berkeley Robot Learning Lab and co-directs BAIR; his honors include the ACM Prize in Computing and IEEE Fellow status.

Noam Brown is an AI researcher at OpenAI, listed on his homepage as working on reasoning, reinforcement learning, self-play and multi-agent AI. He is known for the poker systems Libratus and Pluribus and for work on CICERO and reasoning models such as OpenAI o1.

Princeton University, chartered in 1746 as the College of New Jersey, is where Turing, McCarthy and Minsky took their doctorates and where Fei-Fei Li began ImageNet; seventeen Turing Award laureates have been affiliated with it.

A public research university in Manchester, England, which dates its founding to 1824 and took its present form in 2004; the Manchester Baby, the first electronic stored-program computer, was built there.

University College London (UCL) is a public research university in central London. Geoffrey Hinton led its Gatsby Computational Neuroscience Unit, and Demis Hassabis, a Gatsby postdoctoral researcher, later co-founded DeepMind.

New York University, chartered in 1831 and centred on Washington Square Park, is a private research university whose Courant Institute and Center for Data Science made it a home for machine learning research and data science training.

EPFL, the École Polytechnique Fédérale de Lausanne, is a technical university founded in 1853 and renamed in 1969; it organizes its work around three missions: teaching, research and innovation.

A private research university in Ithaca, New York, founded in 1865. It is also where Frank Rosenblatt studied and worked, and where the perceptron — an early learning machine and the first neural network — was developed.

A public technical university in Munich, founded in 1868, where Sepp Hochreiter wrote the 1991 thesis that produced the first LSTM results, and where Jürgen Schmidhuber studied and later taught.

A French-language public research university in Montreal, founded in 1878 and autonomous since 1919, known today as a centre of deep-learning research and the home institution of Yoshua Bengio and Mila.

A public research university in Beijing's Haidian district, founded in 1911 as Tsinghua College. It is now a major centre for computer science and AI research, with which Turing laureate Andrew Yao, ResNet author Kaiming He and spin-out Zhipu AI are associated.

Zhang Bo (born 1935) is a scholar in automatic control and artificial intelligence. He studied and taught at Tsinghua University and helped found the State Key Laboratory of Intelligent Technology and Systems.

A society for computing educators, researchers and professionals, founded in 1947 at Columbia University and headquartered in New York, with nearly 110,000 members and the Turing Award among its recognitions.

IBM's first electronic computer, announced to shareholders in April 1952. Built as a general-purpose machine, it ran some 25 to 50 times faster than its predecessors and opened IBM's 700 series.

HHMI is a private biomedical research institution founded in 1953 that funds scientists at all career stages and runs the Janelia Research Campus.

k-means clustering splits n observations into k groups by repeatedly sending each point to its nearest centroid and recomputing each centroid as its cluster's mean. It is simple and scales to large data, but k must be fixed in advance.

Yoav Shoham is an Israeli-American computer scientist, professor emeritus at Stanford and co-founder and chairman of AI21 Labs. He initiated the AI Index and has worked on AI, game theory and multi-agent systems.

A public research university in La Jolla, San Diego, founded in 1960, where connectionist neural-network research flourished in the 1980s and where data science and computing were later organised into dedicated institutes and a school.

Tang Xiaoou (1968–2023) was a Chinese computer vision researcher at the Chinese University of Hong Kong and co-founder and chairman of SenseTime, one of China's best-known AI companies.

A model validation technique that repeatedly splits data into folds, trains on some parts and validates on others, then averages the results to estimate how well a model or statistical result will generalise to unseen data.

AAAI is an international nonprofit scientific society for artificial intelligence, founded in 1979, which runs a flagship annual conference and has published a quarterly magazine since 1980.

Dragon Systems was a speech recognition company in Newton, Massachusetts, founded in 1982 by James K. Baker and Janet M. Baker. It made DragonDictate and NaturallySpeaking; its product line later passed to Nuance.

Ali Farhadi (born 1982) is a computer scientist and professor at the University of Washington who served as chief executive of the Allen Institute for AI from 2023 to 2026, a period that included fully open releases of the OLMo, Molmo and Tülu models.

Spanish computer scientist (born 1983) who co-invented sequence-to-sequence learning, led DeepMind's AlphaStar research group, and became a technical lead on Gemini before leaving to start Discovery Loop.

A Swiss artificial intelligence institute founded in 1988 and now a joint institute of USI and SUPSI, known for LSTM and other neural-network research.

Wojciech Zaremba is a Polish computer scientist who co-founded OpenAI in 2015. He later led teams working on GPT models, GitHub Copilot and Codex.

The UCI Letter Recognition dataset is an archive of 20,000 distorted capital-letter images, each reduced to 16 integer attributes, used to benchmark classification algorithms on a 26-class recognition task.

Honda's P1 was the first humanoid in Honda's P series, built during 1993–1997 research. Standing 1,915 mm and weighing 175 kg, it combined bipedal legs with a torso and arms, running on external power and an external computer.

A 1993 journal article by Mitchell P. Marcus, Beatrice Santorini and Mary Ann Marcinkiewicz describing the Penn Treebank: over one million words of Wall Street Journal text annotated with syntactic bracketing, built as a resource for work on English sentence structure.

Honda's P3, completed in September 1997, is a completely independent bipedal humanoid walking robot, 1,600 mm tall and 130 kg; Honda later called ASIMO its further evolved version.

Microsoft Research Asia is Microsoft's research laboratory in the Asia-Pacific region, founded in Beijing in November 1998. It works on artificial intelligence, human-computer interaction and related fields, and its researchers have included the first author of the ResNet paper.

Founded in 2000 as the Singularity Institute, MIRI is a Berkeley nonprofit that was among the first to treat artificial intelligence risk as a technical research problem, and now advocates halting work on smarter-than-human AI.

Graph neural networks are neural networks built for graph-structured data, in which nodes repeatedly exchange messages with their neighbours. Since the model was formulated in 2009, they have become a standard way of learning on graphs.

Megvii is a Chinese computer vision company founded in 2011 by three Tsinghua graduates. Its Face++ cloud platform opened face recognition to developers, and it was the earliest founded of the "CV four little dragons."

DNNresearch Inc. was a University of Toronto start-up founded in 2012 by Geoffrey Hinton and two of his graduate students; Google bought it for its deep neural network research.

A regularisation technique for neural networks: during training it randomly zeroes the inputs or outputs of some neurons so that units cannot co-adapt, which reduces overfitting.

Yitu Technology is a Chinese computer vision company founded in September 2012 by Zhu Long and Lin Chenxi. It is counted among the "CV four little dragons", alongside Megvii, SenseTime and CloudWalk, that grew out of computer vision in China's early AI industry.

A 2013 paper by Goodfellow, Warde-Farley, Mirza, Courville and Bengio introduces maxout units, whose output is the maximum of a set of inputs, designed to work with dropout, and reports state-of-the-art results on four benchmarks.

A 2013 paper by Graves, Mohamed and Hinton that examines deep recurrent neural networks for speech recognition, arguing for end-to-end sequence training with methods such as connectionist temporal classification.

NVIDIA's Tesla K80 is a dual-GPU data-center accelerator announced in November 2014, putting two GK210 chips, 24 GB of memory and up to 8.74 teraflops of single-precision performance on one 300 W board.

A 2014 paper by Mehdi Mirza and Simon Osindero that adds conditioning data to both the generator and the discriminator, enabling a GAN to generate samples such as MNIST digits conditioned on class labels.

A 2015 arXiv paper by Kelvin Xu, Jimmy Ba, Ryan Kiros, Kyunghyun Cho and co-authors that introduced an image-captioning model with visual attention, able to show where it is looking while writing each word.

CloudWalk Technology is a Chinese AI company built on academic face recognition research. Founded in April 2015, it was the last of the four Chinese computer vision firms known as the "CV four little dragons" to be established.

A nonprofit coalition of industry, academic, civil society and media organizations, formed in 2016 to guide the responsible development and use of artificial intelligence.

Handle was a Boston Dynamics research robot that balanced on two wheels and used Atlas-derived arms to move objects; the company credits it with truck unloading and pallet building, and says the work led to its warehouse robot Stretch.

An independent, non-profit AI research institute in Toronto, launched in 2017 with provincial and corporate funding, with Geoffrey Hinton among its founding scientists.

The AI Now Institute is an American research institute, launched in 2017 in partnership with New York University, that studies the social implications of artificial intelligence and the concentration of power in the tech industry.

Group Normalization is a 2018 paper by Yuxin Wu and Kaiming He that proposes normalizing channels in groups, independent of batch size, to address batch normalization's weakness when batches are small.

A 2018 paper by Diederik P. Kingma and Prafulla Dhariwal presenting Glow, a flow-based generative model built on invertible 1x1 convolutions, which improved log-likelihood on standard benchmarks and could synthesise and manipulate large images.

Stanford's Institute for Human-Centered Artificial Intelligence, launched in 2019 and co-directed by Fei-Fei Li and John Etchemendy, studies AI's opportunities and risks and houses the Center for Research on Foundation Models.

HellaSwag is a 2019 benchmark in which a machine must pick the most plausible continuation of a short event description. Its wrong answers are generated and filtered to fool models rather than people.

ERNIE 2.0 is Baidu's pre-trained language model framework, published in 2019, which builds pre-training tasks incrementally through continual multi-task learning and reported results above BERT and XLNet on English and Chinese benchmarks.

An AI processor from Cerebras Systems that keeps an entire silicon wafer as one chip, holding more than 1.2 trillion transistors and 400,000 cores; the first of a line that reached WSE-3 in 2024.

MuZero is a DeepMind reinforcement-learning algorithm that pairs tree-based search with a learned model. It reached superhuman play in Atari, Go, chess and shogi without being given the environment's rules.

Meena is a 2.6-billion-parameter open-domain chatbot from Google, trained end-to-end on filtered social media conversations. Its 2020 paper introduced the SSA human-evaluation metric and linked next-token perplexity to human-judged chat quality. The work was later renamed LaMDA.

Epoch AI is a data-first research nonprofit studying the future of artificial intelligence, known for tracking AI training-compute trends, hosting databases on hardware and compute, and running independent model evaluations and its FrontierMath benchmark.

GPT-Neo is a family of open text-generation models released by EleutherAI in March 2021 at 125M, 1.3B and 2.7B parameters, built to replicate the GPT-3 architecture and distributed with freely downloadable weights under an MIT license.

Gopher is DeepMind's 280-billion-parameter language model, announced in December 2021 alongside a study of what scale does to language models and a companion taxonomy of their risks.

ERNIE 3.0 Titan is a Chinese pre-trained language model with over 260 billion parameters, described by its authors as the largest dense Chinese model of its time and reported to beat state-of-the-art systems on 68 NLP datasets.

A San Francisco nonprofit founded in 2022 that researches technical AI safety, builds the field of safety researchers, and advocates for safety standards — known for a 2023 statement on AI extinction risk signed by industry leaders.

AlphaCode is a DeepMind code-generation system for competitive programming, announced in February 2022. On Codeforces contests held after its training data, it reached an estimated rank around the median participant. A Gemini-based AlphaCode 2 followed in December 2023.

Flamingo is a DeepMind visual language model announced in April 2022. It takes prompts that mix images, videos and text and replies in language, learning new tasks from only a few examples.

OPT-175B is a 175-billion-parameter language model from Meta AI, announced in May 2022. Meta said it was the first model of its size released with both pretrained weights and the code to train and use them.

Gato is a generalist agent from DeepMind, announced on 12 May 2022. A single network with one set of weights handles text, images, Atari games and robot control, aiming at multi-task, multi-embodiment learning.

Galactica is a large language model for science, unveiled by Meta AI and Papers with Code in November 2022. Its public demo was taken down about three days later, after users found it produced fake papers.

Pythia is a suite of 16 open language models from EleutherAI, ranging from 70 million to 12 billion parameters, all trained on the same data in the same order, with 154 public checkpoints from each training run.

A Google DeepMind generative world model with 11 billion parameters that turns text, images or sketches into controllable virtual worlds, described as the first generative interactive environment trained on unlabelled internet videos without action labels.

OpenELM is a family of small open language models from Apple, published in four sizes from 270M to 3B parameters with its training framework, logs, checkpoints and MLX conversion code.

Cinematch was Netflix's own algorithm for predicting what rating a subscriber would give a film. Its accuracy became the baseline of the Netflix Prize, a $1 million contest to beat it by 10 percent.

Shane Legg is a New Zealand AI researcher and co-founder of DeepMind, known for helping popularise the term "artificial general intelligence" and for leading the company's AGI safety work.

Xiaodong He is a JD.com vice president, deputy managing director of JD AI Research, and an IEEE Fellow. He works on conversational AI, including question answering, emotion detection, and public-service hotlines.

Zhiyuan Liu is a tenured associate professor at Tsinghua University and co-founder and chief scientist of Beijing ModelBest, a company working to run capable large models on local devices.

AI researcher and research administrator who worked at IBM's Watson Group and at JD.com, and has been director and chief scientist of the Shanghai AI Laboratory since April 2024.

Decision tree learning builds a classification or regression tree from data by repeatedly splitting it into more homogeneous groups. It is a widely used supervised method whose rules can be read directly, and whose main weaknesses are instability and a tendency to overfit.

Ziv Aviram is a co-founder of Mobileye, the company he helped build from its founding in 1999 through its 2014 New York Stock Exchange listing and its 2017 acquisition by Intel.

Soccer Server is a software simulator of soccer, begun by Itsuki Noda as a platform for multi-agent systems research, which later became the official soccer server of RoboCup.

Jack Clark is a co-founder of Anthropic, where he has led policy and later public-benefit work and, from March 2026, the Anthropic Institute. He is also a co-author of the GPT-3 paper.

XRCE's computer-vision team competed in the ImageNet challenge and won the 2011 classification task using high-dimensional image signatures, product quantization and linear SVMs; its Fisher-vector methods performed strongly across the 2010–2014 challenges.

Tang Jie is a Chinese computer scientist listed as a founder of Zhipu AI and its chief scientist, and a Tsinghua professor who created the AMiner academic database and co-authored the GLM pretraining work behind Zhipu's models.

An autoencoder is an unsupervised neural network that compresses its input into a low-dimensional code and reconstructs it back, learning a latent representation useful for dimensionality reduction, denoising and anomaly detection.

Sun Jian was a computer-vision scientist who spent 13 years at Microsoft, co-created ResNet, and later became chief scientist at Megvii. He died aged 45 after a sudden illness in June 2022.

Shun-ichi Amari (born 1936) is a Japanese mathematical engineer who founded information geometry and built theoretical foundations for machine learning and neural networks; he received the 2025 Kyoto Prize in Advanced Technology.

Chris Urmson is co-founder and chief executive officer of Aurora, the self-driving company he leads. Under him Aurora announced the acquisition of Uber's self-driving unit and put driverless commercial trucks on public roads.

Tsinghua University's Institute for AI Industry Research (AIR) is an applied AI research institute set up under Zhang Yaqin, targeting the fourth industrial revolution through autonomous driving, AI plus IoT and brain-inspired intelligence.

BIGAI is a non-profit AI research institute in Beijing, led by Zhu Songchun, that works with Peking and Tsinghua universities toward a unified theory of general intelligence and general intelligent agents.

Peking University is a Chinese university that dates its founding to 1898 and reported 53,946 full-time students in a January 2026 snapshot; in September 2020 Zhu Songchun was named to lead its Institute for Artificial Intelligence.

A large comprehensive research university in China, with 47 schools, 16 disciplines on the national Double First-Class list, more graduate than undergraduate students, and 36 members of the national academies on its staff.

Zhang Yaqin is an electrical engineer and technology executive who spent nearly 16 years at Microsoft, served as president of Baidu from 2014 to 2019, and joined Tsinghua University in 2019 as a chair professor of intelligent science.

Sam McCandlish is a co-founder and chief technology officer of Anthropic, a co-author of influential work on scaling laws and GPT-3, and the company's first Responsible Scaling Officer.

Kyle Vogt is identified in General Motors' March 2016 announcement as the founder of Cruise Automation, the San Francisco startup GM acquired; the company he founded later ran driverless passenger service in San Francisco until California regulators suspended its permits in October 2023.

UCLA is the University of California's Los Angeles campus. Its own figures show a large, heavily undergraduate institution: 46,678 students in total, most of them Californians, taught by 5,464 faculty.

Zhu Songchun is a computer vision researcher who worked at the University of California, Los Angeles from 2002 and was named in 2020 to lead Peking University's Institute for Artificial Intelligence.

Redwood Research is an AI safety research group that has co-authored, with Anthropic, studies of deceptive behaviour in language models — sleeper-agent backdoors and alignment faking — as well as work on AI control protocols.

Tom Brown is an AI researcher listed among the co-founders of Anthropic and the first author of the 2020 paper that introduced GPT-3, one of the most widely cited demonstrations of few-shot learning in large language models.

First author of the latent diffusion paper that underlies Stable Diffusion, and later co-founder and CEO of Black Forest Labs, the company behind the FLUX image models.

Andreas Blattmann is a researcher on generative image models, second author of the latent diffusion paper behind Stable Diffusion, co-author of the Stable Diffusion 3 paper, and a 2024 co-founder of Black Forest Labs, which released the FLUX.1 models.

Patrick Esser is a researcher on generative image and video models: a co-author of the latent diffusion paper that became the basis of Stable Diffusion, lead author of the Stable Diffusion 3 paper, and a 2024 co-founder of Black Forest Labs.

Wang Haifeng is Baidu's chief technology officer, an ACL and IEEE Fellow, and the first-listed author of the ERNIE technical reports behind the company's foundation models.

Jacob Devlin is a research scientist and first author of BERT, a language model Google introduced in 2018 that reads text in both directions. He earlier worked on neural machine translation at Microsoft Research.