Top 100 AI Researchers Who Communicate

102people

Andrej Karpathy

"Software 2.0", neural net zero-to-hero lecture series, nanoGPT

GANs; co-author of the Deep Learning textbook

"The Bitter Lesson"; reinforcement learning textbook

deep learning foundations; AI risk essays

DeepMind founder; AlphaGo/AlphaFold communication

"Machines of Loving Grace"; Anthropic scaling essays

scaling hypothesis advocacy; seminal lectures on representation

convolutional nets; JEPA / world-model essays

Distill.pub founder; mechanistic interpretability primers

Geoffrey Hinton

backprop, capsules; public AI-risk communication

Coursera ML course; "AI is the new electricity"

ImageNet; "The Worlds I See"; human-centered AI framing

Sebastian Raschka

"Build a Large Language Model from Scratch"

long-form ML blog (lilianweng.github.io)

"Illustrated Transformer", "Illustrated GPT-2"

Annotated Transformer; minGPT-style pedagogy

fast.ai courses; ULMFiT

fast.ai co-founder; AI ethics essays

"Human Compatible"; provably beneficial AI

"The Master Algorithm"; five tribes framework

Michael Nielsen

"Neural Networks and Deep Learning" online book

"Information Theory, Inference, and Learning Algorithms"

Christopher Bishop

"Pattern Recognition and Machine Learning"

"Probabilistic Machine Learning" textbooks

Jurgen Schmidhuber

LSTM; deep history of deep learning essays

Berkeley RL lectures; Covariant founder essays

Berkeley deep RL course; offline RL writing

meta-learning (MAML); Stanford CS330 lectures

HELM benchmark; foundation-model framing

Christopher Manning

Stanford NLP course; CS224n

"Speech and Language Processing" textbook

"Stochastic Parrots"; octopus-test framing

algorithmic-bias research; DAIR Institute

Margaret Mitchell

Model Cards framework

Arvind Narayanan

"AI Snake Oil"; CS-fairness writing

"AI Snake Oil" co-author

Anthropic Interpretability Team

circuits, sparse autoencoder, monosemanticity reports

mechanistic interpretability tutorials and TransformerLens

Jacob Steinhardt

emergent capabilities essays; Berkeley alignment

Paul Christiano

RLHF; alignment-research essays

superalignment writing; OpenAI/Anthropic

Holden Karnofsky

"Most important century" series

biological-anchors timelines report

Eliezer Yudkowsky

"Sequences"; AI-risk arguments

long-form scaling-law and tooling essays

Interconnects; RLHF book

Sebastian Ruder

NLP newsletter; transfer-learning surveys

paper-explained YouTube channel

"Grokking Deep Learning"; OpenMined

François Chollet

Keras; ARC benchmark; "On the Measure of Intelligence"

Aurélien Géron

"Hands-On Machine Learning with Scikit-Learn & TensorFlow"

Georgia Tech ML lectures; community essays

Michael Littman

RL lectures; "code that runs other code" framing

canonical ML textbook; "well-posed learning problem"

probabilistic graphical models textbook

causal inference; "The Book of Why"

interpretable-ML manifesto

TCAV; concept-based interpretability

Finale Doshi-Velez

interpretable-ML rigor framework

"Troubling Trends in ML Scholarship"

"Machine learning is alchemy" NeurIPS test-of-time talk

"Dive into Deep Learning" book

D2L co-author; distributed-training essays

bitsandbytes; GPU/LLM hardware blog

Hugging Face team (Thomas Wolf et al.)

Transformers library tutorials and ecosystem essays

"NLP with Transformers" book

Leandro von Werra

TRL library; RLHF tutorials

Sebastian Bubeck

"Sparks of AGI" report; theory blog

NLP textbook; LLM critique threads

Hadley Wickham (adjacent)

tidymodels; data-science framework writing

"Think Bayes", "Think Stats"

Cassie Kozyrkov

decision-intelligence framework essays

"Designing Machine Learning Systems"; LLM-eval blog

applied ML at scale blog

embeddings primer; "What are embeddings?"

LLM eval and fine-tuning playbooks

ML pedagogy blog

Christopher Potts

Stanford NLU; benchmark-design essays

Stephen Wolfram

"What Is ChatGPT Doing…" essay

multi-task learning; intelligible models

"Pathways" essay; Google AI architecture writing

seq2seq; AlphaStar communications

RL course at UCL; AlphaGo lectures

reproducibility-in-RL framework; Meta FAIR

multi-agent RL primers

Mixture-of-Experts, multi-query attention papers (and explainers)

Cohere co-founder; transformer-builder commentary

scaling-laws paper

GPT-3 paper lead; few-shot framing

GPT-1/2/CLIP communications

PPO; RLHF lecture notes

World Models paper; Sakana essays

efficient-Transformers survey; long-context blog

FlashAttention papers and explainers

Aleksander Madry

adversarial-robustness framework

secure ML and decentralized-AI writing

assistive-AI / human-robot interaction lectures

"stochastic relaxation" foundations; pedagogical writing

Michael I. Jordan

graphical models; "AI revolution hasn't happened yet" essay

Cohere For AI; "The Hardware Lottery" essay

OpenAI researcher and lead on o1 (inference-time scaling / chain-of-thought reasoning). Co-created Libratus (2017 poker AI). Regularly public-speaking on the case that inference-time compute is a new scaling axis.

NVIDIA Director of Robotics; created Voyager (open-ended LLM agent) and coined the "Physical Turing Test". Co-leads GEAR Lab + Project GR00T on embodied / physical AI.