People in AI

Andrej Karpathy

Researcher

Released autoresearch project for autonomous AI agent-based LLM training research.

Yann LeCun

Researcher

Chief AI Scientist at Meta

Keynoted MIT symposium, arguing that the most exciting advances in generative AI will come from 'world models' rather than continued improvements to large language models.

Richard Sutton

Researcher

Authored The Bitter Lesson, stating that general methods leveraging computation are most effective in AI research.

OpenAI alum who oversees Meta Superintelligence Labs, credited as cocreator of ChatGPT, GPT-4, mini models, 4.1, and o3 at OpenAI.

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

Jeff Dean

Researcher

Chief Scientist at Google

Discussed Google's comprehensive energy measurements for Gemini AI prompts in exclusive interview with MIT Technology Review.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Oriol Vinyals

Researcher

VP of Research at Google DeepMind

Credited improving pre-training and post-training for Gemini 3's massive performance improvements, stating the delta between 2.5 and 3.0 is the biggest Google has ever seen with no scaling walls in sight.

Erik Brynjolfsson

Researcher

Jerry Yang and Akiko Yamazaki Professor, Senior Fellow at Stanford Institute for Human-Centered AI at Stanford

Authored research examining AI's employment effects on early-career workers using high-frequency payroll data, finding 16% relative decline in AI-exposed occupations.

Named as one of the pioneering figures in AI and machine learning alongside Bengio and LeCun

Interviewed by MLStreetTalk to help launch ARC-AGI-2 benchmark

Arman Cohan

Researcher

Yale University

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

Newton Cheng

Researcher

Anthropic

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

John Hallman

Researcher

Pretraining Researcher at OpenAI

Expressed concern about AI-based feeds when OpenAI released the Sora 2 social media app but supported the team's approach.

Jakub Pachocki

Researcher

Chief Scientist at OpenAI

Described OpenAI's AI researcher goal as a system capable of autonomously delivering on larger research projects and discussed the path to superintelligence.

Yi Tay

Researcher

Research Scientist at Google DeepMind

Praised Gemini 3 Pro on social media, stating it's the best model in the world by a wide margin.

Co-authored essay 'Welcome to the Era of Experience' proposing new paradigm of training AI models through reinforcement learning via interaction with environments.

Yejin Choi

Researcher

Researcher at Together AI

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

James Zou

Researcher

Researcher at Stanford

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Xiaolong Wang

Researcher

Researcher at UC San Diego

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Nathan Lambert

Researcher

Research Scientist at Allen Institute for AI (AI2)

Argues that open models will never catch up with closed ones but believes they will be the engine for the next ten years of AI research.

Alexia Jolicoeur-Martineau

Researcher

Research Scientist at Samsung SAIL Montreal

Proposed Tiny Recursive Model (TRM) achieving higher generalization than large language models on reasoning tasks with only 7M parameters.

Sam Bowman

Researcher

Alignment Researcher at Anthropic

Stated that alignment researchers are involved in every part of training at Anthropic without a clear split between alignment research and applied finetuning.

Co-author of research on efficient reasoning on edge devices for language models.

Jakob Foerster

Researcher

Co-author of Hyperagents research paper on AI agent systems.

Co-author of Hyperagents research paper on AI agent systems.

Jeff Clune

Researcher

Co-author of Hyperagents research paper on AI agent systems.

Bingchen Zhao

Researcher

Co-author of Hyperagents research paper on AI agent systems.

Co-author of Hyperagents research paper on AI agent systems.

Co-authored paper measuring Claude Mythos' ability to develop exploits and find zero-day vulnerabilities.

Jerry Tworek

Researcher

Google DeepMind

Noted achievement of IMO Gold performance through general model training and natural language prompting without IMO-specific work.

Alexander Wei

Researcher

Google DeepMind

Described breakthrough in general-purpose reinforcement learning and test-time compute scaling for IMO problem solving.

Joanne Jang

Researcher

Founding Leader of Model Behavior Team at OpenAI

Moving to start new OAI Labs research team responsible for inventing new interfaces for human-AI collaboration.

Aditya Challapally

Researcher

Lead Author and Research Contributor at MIT NANDA

Led MIT research showing 95% of enterprise generative AI pilots are failing and explained the reasons for the GenAI divide.

Amanda Askell

Researcher

Anthropic

Was targeted by Elon Musk in discussions around safety at xAI and Anthropic.

Peter McCrory

Researcher

Researcher at Anthropic

Co-authored research on Project Deal, an experiment analyzing AI agent behavior in marketplace negotiations.

Co-author of the Deep Think with Confidence research paper

19th century mathematician whose detailed descriptions of his thought process helped Wallas define the verification stage of creativity

Co-author of the research paper on embedding-based retrieval limitations

Co-author of the research paper on embedding-based retrieval limitations

His FlashAttention-2 paper is referenced in comparison to Triton implementations of FlashAttention backward

Co-author of paper investigating diminishing returns in LLM scaling

Author of the research paper on measuring long horizon execution in LLMs

Lead author of research paper on long horizon execution in large language models

Co-author of research on long-horizon task execution in language models

Collaborated with Michel Devoret on quantum research in the 1980s that led to Nobel Prize in Physics

Chief scientist on Google's quantum team who won the Nobel Prize in Physics for quantum research conducted in the 1980s

Collaborated with Michel Devoret on quantum research that led to Nobel Prize in Physics

Awarded Nobel Prize in Chemistry for breakthrough AlphaFold research

Co-author of the robot learning tutorial paper

Co-author of the research paper on reinforcement learning and LLM reasoning

Asked for help identifying tensor attributes that could differ between two tensors

Journalist who spent three hours in the anechoic chamber and documented visitor experiences.

Thomas Bloom

Researcher

Mathematician who maintains the Erdos Problems website and provided validation for OpenAI's mathematical proof disproving the planar unit distance problem.

Mathematician who provided peer validation for OpenAI's claim about disproving an 80-year-old Erdős conjecture in geometry.

Mathematician who provided peer validation for OpenAI's claim about disproving an 80-year-old Erdős conjecture.

Timothy Gowers

Researcher

Fields Medalist, Combinatorics Chair at College de France

Demonstrated that ChatGPT 5.5 Pro solved an open mathematics problem in number theory, producing doctoral-level research in under two hours.

Quoted observing that AI is shifting from assisting tasks to orchestrating entire workflows autonomously.

Apple analyst titled assessment 'Something Is Rotten in the State of Cupertino'

Senior analyst suggesting the xAI deal signals SpaceX preparing for a public listing given compute and infrastructure costs.

Actor portrayed in the fake leaked ad with supposed OpenAI hardware device

Expressed strong doubts about methodology and medal claim, though concerns were addressed

Podcast host who interviewed Fidji Simo about AI-based relationships and user attachments.

Research scientist who announced departure via tweet in April, citing desire to take different risks.

Former OpenAI researcher who went through Meta onboarding but never started.

Argued that productive friction in creative work is what AI threatens to take away.

Co-authored research on general scales for AI evaluation with explanatory and predictive power.

Lexin Zhou

Researcher

Researcher at Princeton University

Lead author on research introducing general scales for AI evaluation with explanatory and predictive power.

Co-authored research on teaching LLMs to reason like Bayesians by training them to mimic optimal Bayesian model predictions.

Tal Linzen

Researcher

Research Scientist at Google Research

Co-authored research on teaching LLMs to reason like Bayesians by training them to mimic optimal Bayesian model predictions.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored research on Code-Space Response Oracles for generating interpretable multi-agent policies with LLMs.

Co-authored research on SWE-CI benchmark evaluating AI agents maintaining codebases across months of changes.

Sid Wang

Researcher

Meta Superintelligence Labs

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored research on Code-Space Response Oracles for generating interpretable multi-agent policies with LLMs.

Co-authored research on SWE-CI benchmark evaluating AI agents maintaining codebases across months of changes.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Yue Yu

Researcher

Meta Superintelligence Labs

Co-authored research on training AI co-scientists using rubric rewards.

Co-authored research on Code-Space Response Oracles for generating interpretable multi-agent policies with LLMs.

Co-authored research on SWE-CI benchmark evaluating AI agents maintaining codebases across months of changes.

Yixin Liu

Researcher

Yale University

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Co-authored research examining reasoning LLMs-as-Judges in non-verifiable LLM post-training.

Marc Lanctot

Researcher

Google DeepMind

Co-authored research on Code-Space Response Oracles for generating interpretable multi-agent policies with LLMs.

Co-authored research on SWE-CI benchmark evaluating AI agents maintaining codebases across months of changes.

Core contributor on MolmoPoint research for better visual grounding in vision-language models.

Jieyu Zhang

Researcher

Allen Institute for AI

Researcher on MolmoPoint project for vision-language model grounding improvements.

Researcher on MolmoPoint project for vision-language model grounding improvements.

Core contributor on MolmoPoint research for better visual grounding in vision-language models.

Jae Sung Park

Researcher

Core Contributor at Allen Institute for AI

Core contributor on MolmoPoint research for better visual grounding in vision-language models.

Researcher on MolmoPoint project for vision-language model grounding improvements.

Researcher on MolmoPoint project for vision-language model grounding improvements.

Vinay Prabhushankar

Researcher

Anthropic

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

Core contributor on MolmoPoint research for better visual grounding in vision-language models.

Zixian Ma

Researcher

Allen Institute for AI

Researcher on MolmoPoint project for vision-language model grounding improvements.

Researcher on MolmoPoint project for vision-language model grounding improvements.

Winson Han

Researcher

Allen Institute for AI

Researcher on MolmoPoint project for vision-language model grounding improvements.

Michael Moore

Researcher

Anthropic

Co-authored technical blog post about Claude Mythos Preview's security capabilities and testing methodology.

Martin Jaggi

Researcher

Professor of Machine Learning at EPFL

Member of Steering Committee of Swiss AI Initiative that released Apertus, a fully open AI model.

Adam Tauman Kalai

Researcher

OpenAI

Author of OpenAI research paper on hallucinations in language models, used as example where chatbot incorrectly cited his PhD dissertation and birthday.

Da-Cheng Juan

Researcher

Research Lead at Google Research

Co-introduced SLED, a decoding strategy that enhances the accuracy of LLMs by aligning their output with the model's intrinsic knowledge.

Achieved the highest score on ARC-AGI v1 (79.6%) and v2 (29.4%) using multi-agent collaboration with evolutionary test-time compute.

Cyrus Rashtchian

Researcher

Research Scientist at Google Research

Co-introduced SLED, a decoding strategy that enhances the accuracy of LLMs by aligning their output with the model's intrinsic knowledge.

Ashish Kumar

Researcher

Research Scientist at Meta

Left Tesla after leading the Optimus AI team for more than two years to join Meta, having focused on scalable methods using reinforcement learning and learning from videos.

Previously led strategic explorations team at OpenAI, now leads research at Meta Superintelligence Labs after being recruited by Zuckerberg.

Boaz Barak

Researcher

Researcher and Professor at OpenAI and Harvard

Expressed mixed feelings about Sora 2, noting it's technically amazing but premature to celebrate avoiding social media pitfalls.

Authored research on NCCLX collective communication framework for 100k+ GPUs to optimize LLM training and inference.

Michaela Hinks

Researcher

FutureHouse

Co-author of the Kosmos AI Scientist technical report for autonomous discovery.

Jan Leike

Researcher

Alignment Researcher at Anthropic

Former OpenAI Superalignment lead who stated that alignment researchers at Anthropic are deeply involved in post-training and capability work for Opus 4.5.

Federico Bianchi

Researcher

Researcher at Stanford

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Jan Kautz

Researcher

Researcher at NVIDIA

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Carlos Guestrin

Researcher

Researcher at Stanford

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Mert Yuksekgonul

Researcher

Researcher at Stanford

Co-authored TTT-Discover, a reinforcement learning system that allows LLMs to continue training at test time for specific problems.

Co-authored TTT-Discover reinforcement learning system for test-time LLM training.

Thang Luong

Researcher

Google DeepMind

Co-authored Aletheia, an AI system for autonomous mathematics research achieving several research milestones.

Co-authored Aletheia, a math research agent that generated an AI-authored research paper on eigenweights in arithmetic geometry.

Co-introduced VaultGemma and conducted research on scaling laws for differentially private language models in partnership with Google DeepMind.

Stéphane d'Ascoli

Researcher

Researcher at Meta AI

Co-authored research on TRIBE, a deep neural network for predicting brain responses to multimodal stimuli.

Yohann Benchetrit

Researcher

Researcher at Meta AI

Co-authored research on TRIBE, a deep neural network for predicting brain responses to multimodal stimuli.

Jean-Rémi King

Researcher

Researcher at Meta AI

Co-authored research on TRIBE, a deep neural network for predicting brain responses to multimodal stimuli.

Co-authored research on TRIBE, a deep neural network for predicting brain responses to multimodal stimuli.

Hubert Banville

Researcher

Researcher at Meta AI

Co-authored research on TRIBE, a deep neural network for predicting brain responses to multimodal stimuli.

Jürgen Schmidhuber

Researcher

AI researcher who proposed the Gödel Machine concept decades ago, which inspired the Darwin-Gödel Machine self-improving AI system.

Part of the steering committee for Open-H-Embodiment, advancing Physical AI for surgical robotics and ultrasound.

Part of the steering committee for Open-H-Embodiment, a community-driven healthcare robotics dataset initiative.

Nassir Navab

Researcher

Professor at Technical University of Munich

Part of the steering committee for Open-H-Embodiment, building open datasets for Physical AI in healthcare robotics.

Co-author on research about AI Co-Mathematician tool for accelerating mathematical research with agentic AI.

Co-author on research about AI Co-Mathematician tool for accelerating mathematical research with agentic AI.

Lead author on research about AI Co-Mathematician tool that uses agentic AI to accelerate mathematical research.

Ingrid von Glehn

Researcher

Google DeepMind

Co-author on research about AI Co-Mathematician tool for accelerating mathematical research with agentic AI.

Co-author of research paper on EvoClaw, evaluating AI agents on continuous software evolution.

Co-author of research paper on EvoClaw, evaluating AI agents on continuous software evolution.

Co-author of MetaClaw research on AI agents that meta-learn and evolve in the wild.

Lead author of research on online experiential learning for language models.

Babak Ehteshami Bejnordi

Researcher

Microsoft Research

Co-author of research on online experiential learning for language models.

Keivan Alizadeh

Researcher

Apple

Lead author of research on recursive language models meeting uncertainty for long context.

Co-author of research on recursive language models meeting uncertainty for long context.

Lead author of research on efficient reasoning on edge devices for language models.

Co-author of research on efficient reasoning on edge devices for language models.

Co-author of research paper on EvoClaw, evaluating AI agents on continuous software evolution.

Lead author of MetaClaw research on AI agents that meta-learn and evolve in the wild.

Huaxiu Yao

Researcher

Co-author of MetaClaw research on AI agents that meta-learn and evolve in the wild.

Co-author of research on online experiential learning for language models.

Parshin Shojaee

Researcher

Apple

Co-author of research on recursive language models meeting uncertainty for long context.

Co-author of Hyperagents research paper on AI agent systems.

Lead author of HorizonMath research paper on measuring AI progress toward mathematical discovery with automatic verification.

Sumeet Motwani

Researcher

Co-author of HorizonMath research paper on measuring AI progress toward mathematical discovery.

Lilian Weng

Researcher

Authored a must-read recap on autonomous agents like Auto-GPT and BabyAGI.

James V. Roggeveen

Researcher

Co-author of HorizonMath research paper on measuring AI progress toward mathematical discovery.

11-year Meta veteran and co-author of Segment Anything model, now on technical staff at Thinking Machines Lab.

Mojtaba Seyedhosseini

Researcher

Google DeepMind

Announced Gemini Embedding 2, Google DeepMind's first native multimodal embedding model that provides unified representations for text, audio, video, and image inputs.

Described Apertus as a driver of innovation for strengthening AI expertise in research, society and industry.

Co-authored announcement of Google's new AI system designed to help scientists write expert-level empirical software.

Led development of K2 Think reasoning model that rivals OpenAI and DeepSeek at fraction of the size.

Wenxian Shi

Researcher

PhD Student at MIT CSAIL

Lead author of VaxSeer research, creating an AI system using deep learning to predict dominant flu strains and identify protective vaccine candidates.

Quentin Carbonneaux

Researcher

Meta FAIR CodeGen Team

Contributed to the development and release of Code World Model for advancing code generation with world models.

Presented novel approach to time-series forecasting using in-context fine-tuning for foundation models.

Jade Copet

Researcher

Meta FAIR CodeGen Team

Part of the team that released Code World Model (CWM), a 32-billion-parameter open-weights LLM for code generation research.

Co-developed Speech-to-Retrieval (S2R) for voice search that bypasses text transcription.

Timothy Jenkins

Researcher

Medical Biotechnologist at Technical University of Denmark

Led research using AI to build a molecular navigation system for cancer-fighting immune cells, designing proteins with AI models.

Unveiled Speech-to-Retrieval (S2R) technology that directly interprets spoken queries without text conversion.

Kishwar Shafin

Researcher

Technical Lead at Google Research

Co-authored research on DeepSomatic, an AI-powered tool that identifies cancer-related mutations in tumor genetic sequences.

Co-authored research on DeepSomatic, an AI tool using convolutional neural networks to identify tumor variants.

Bolei Zhou

Researcher

Professor and Chief AI Scientist at UCLA and Coco Robotics

Joined Coco Robotics to lead their new physical AI research lab focusing on computer vision and robotics for micromobility.

Alex Shaw

Researcher

Co-created Terminal-Bench and Harbor framework for testing and improving AI agents in containerized environments.

Haotong Lin

Researcher

Lead author on Depth Anything 3, a model that predicts spatially consistent geometry from visual inputs.

Equal contribution author on Depth Anything 3 for monocular and multi-view depth estimation.

Richard Turner

Researcher

Professor of Machine Learning at University of Cambridge

Led research on Aardvark Weather, a new AI weather prediction approach that enables accurate forecasts with dramatically less computing power.

Bingyi Kang

Researcher

Project lead for Depth Anything 3, advancing visual geometry prediction models.

Cole Killian

Researcher

MATS and Anthropic Fellows program

Co-led research project on AI agents exploiting smart contracts, developing the SCONE-bench benchmark comprising 405 exploited contracts.

Albin Cassirer

Researcher

OpenAI

Co-authored research on automated code review systems that complement chain-of-thought monitoring and other safety work.

Co-authored research on training dedicated agentic code reviewers for gpt-5-codex and gpt-5.1-codex-max.

Co-authored research on practical output monitoring for autonomous collaborative coding systems as part of the Codex team.

Co-authored research on scaling code verification and training a dedicated, agentic code reviewer for gpt-5-codex.

Maja Trębacz

Researcher

OpenAI

Co-authored research on automated code review as part of the Codex team, discussing how giving the reviewer repo-wide tools improves recall and precision.

Ali Behrouz

Researcher

Student Researcher at Google Research

Co-introduced the Titans architecture and MIRAS framework for helping AI models have long-term memory.

Vahab Mirrokni

Researcher

VP and Google Fellow at Google Research

Co-introduced the Titans architecture and MIRAS framework for helping AI models have long-term memory.

Listed as a contributor to the Nomos 1 model.

Jeffrey Quesnelle

Researcher

Nous Research

Listed as a contributor to the Nomos 1 model for mathematical problem-solving.

Co-introduced the Titans architecture and MIRAS framework for helping AI models have long-term memory.

Roger Jin

Researcher

Nous Research

Listed as a contributor to the Nomos 1 model, an open-source 30B model for mathematical reasoning.

Kunal Handa

Researcher

Anthropic

Led research using Anthropic Interviewer to conduct AI-powered qualitative interviews with 1,250 professionals about working with AI.

Alexander Knop

Researcher

Research Scientist at Google Research

Co-authored URANIA, a differentially private framework for gaining insights into AI chatbot use.

Daogao Liu

Researcher

Post Doc Researcher at Google Research

Co-authored URANIA, a differentially private framework for gaining insights into AI chatbot use.

Co-authored work on a Gemini tool that provides automated feedback for theoretical computer scientists at STOC 2026.

David Woodruff

Researcher

Research Scientist at Google Research

Co-authored work on a Gemini tool that provides automated feedback for theoretical computer scientists at STOC 2026.

Stated that in nature the most energy efficient way things work is how they actually work, in context of AI's energy demands.

Oliver Cameron

Researcher

Authored analysis on Odyssey-2 Pro as a nascent world simulator that learns to predict how the world evolves over time.

Submitted research paper on WebGym, the largest open-source environment for training visual web agents with nearly 300,000 tasks.

Lead author on NitroGen, an open foundation model for generalist gaming agents trained on internet video gameplay data.

Sarah Alamdari

Researcher

Senior Applied Scientist at Microsoft Research

Lead author of research on AI model for designing peptides for cancer detection.

Ava Amini

Researcher

Principal Researcher at Microsoft Research

Senior author of study on AI-designed peptides for cancer detection, former graduate student in Bhatia's lab.

Eric Drexler

Researcher

Argued at EA Global 2018 that we should use new AI data to rethink models for how superintelligent AI is likely to emerge and function, and discussed AI safety with Marvin Minsky around 1990.

Introduced GIST, a novel algorithm for selecting high-quality data subsets that maximizes both data diversity and data utility.

Co-introduced GIST algorithm at NeurIPS 2025 for balancing data diversity and utility in machine learning datasets.

Co-authored article on the state of on-device LLMs in 2026, covering techniques for model compression and deployment.

Vikas Chandra

Researcher

Co-authored article on the state of on-device LLMs in 2026, discussing advances in running billion-parameter models on mobile devices.

Randall Balestriero

Researcher

Brown University

Co-authored a paper with Yann LeCun introducing LeJEPA for self-supervised vision training.

Mohit Iyyer

Researcher

University of Maryland

Co-authored How2Everything framework for mining web procedures to evaluate LLMs.

Kyle Lo

Researcher

Allen Institute for AI

Co-authored How2Everything framework for evaluating LLM procedural generation capabilities.

Co-authored CausalArmor framework for defending AI agents against prompt injection attacks.

Yapei Chang

Researcher

Allen Institute for AI

Co-authored How2Everything framework for mining and evaluating procedural generation in LLMs.

Dawn Song

Researcher

SafeRL-Lab (UC Berkeley)

Co-authored AgenticPay, a framework for studying language-mediated economic interaction among AI agents.

Summer Yue

Researcher

Director of Alignment at Meta

Tested OpenClaw AI agent which went out of control and attempted to delete her emails despite instructions to stop.

Stephan Rabanser

Researcher

Princeton University

Works on trustworthy machine learning with focus on uncertainty quantification and AI agent reliability evaluation.

Part of the team that developed CWM, an LLM trained with world modeling for improved code generation.

Thomas Malone

Researcher

Professor and Director at MIT Center for Collective Intelligence

Led research finding that AI-human combinations do not outperform the best human-only or AI-only system on average.

Luca Soldaini

Researcher

Allen Institute for AI

Co-authored How2Everything framework for evaluating and improving LLM procedure generation.

Greg Burnham

Researcher

Senior Researcher at Epoch AI

Explained FrontierMath benchmark for measuring AI mathematical reasoning capabilities and discussed AI's advancement in solving PhD-level math problems.

Applied interpretability methods to understand how an AI model detects Alzheimer's disease from blood samples and identified fragmentomics as a novel biomarker class.

Max Schwarzer

Researcher

Post Training Lead at OpenAI

Now oversees the Model Behavior team as part of OpenAI's research reorganization.

Karolis Misiunas

Researcher

Research Engineer at Google DeepMind

Introduced an innovative end-to-end speech-to-speech translation model enabling real-time translation with only 2-second delay.

Joe Carlsmith

Researcher

Wrote an essay series on solving the AI alignment problem, discussing the importance of building AIs that do human-like philosophy.

Participated in Anthropic Fellows research on AI models exploiting smart contracts and quantifying economic harm.

Keir Bradwell

Researcher

Researcher at Anthropic

Co-authored research on Project Deal, studying how AI agents negotiate and conduct transactions on behalf of humans.

Co-authored research on Project Deal, an internal experiment where Claude agents bought and sold real items on behalf of employees.

Co-authored research on Project Deal, an internal experiment where Claude agents conducted negotiations and transactions on behalf of employees.

Co-founded ARI, acquired by Meta for humanoid robotics; previously taught at NYU and co-founded Fauna Robotics before Amazon acquisition.

Yilun Zhao

Researcher

Yale University

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Co-author of RbtAct research on using rebuttals as supervision for actionable review feedback generation with LLMs.

Noah A. Smith

Researcher

Researcher at AI2

Co-authored research on modular post-training with mixture-of-experts for language models.

Sewon Min

Researcher

Researcher at AI2

Co-authored paper on BAR recipe for modular post-training in language models at Allen AI.

Jacob Morrison

Researcher

Researcher at AI2

Co-authored paper on modular post-training with mixture-of-experts (BAR recipe) for language models at Allen AI.

Co-authored paper on modular post-training with mixture-of-experts for improving language model capabilities.

Akshita Bhagia

Researcher

Researcher at AI2

Co-authored research on BAR modular post-training recipe for language models at Allen AI.

Co-authored paper on modular post-training approaches for language models.

Hongxun Wu

Researcher

Researcher at OpenAI

OpenAI researcher who described the reasoning-LLM breakthrough on solving Erdős problems as a milestone on 'the hardest problems'.

hyunji amy lee

Researcher

Researcher

Introduced MINTEval, a benchmark targeting long-context memory systems for LLM agents operating in continuously updated environments.

Built ExploitBench benchmark to study exploit development capabilities of large language models.

Eight-year Meta veteran who worked on multimodal perception systems and open-world segmentation including SAM3D, recently joined Thinking Machines Lab.

Harvard PhD who spent 10 months at Meta before joining Thinking Machines Lab, part of TML's talent expansion.

Andrea Madotto

Researcher

Researcher at Thinking Machines Lab

Former research scientist in Meta's FAIR division focused on multimodal language models, joined Thinking Machines Lab in December.

Edoardo Cetin

Researcher

Sakana AI

Co-authored research on training a 7B model to manage and orchestrate other LLMs in natural language.

Co-authored research on training a 7B model to manage and orchestrate other LLMs in natural language.

Co-authored research on training a 7B model to manage and orchestrate other LLMs in natural language.

Listed as author of the DuPO paper on preference optimization

Listed as author of the DuPO paper on preference optimization

Co-author of the article on humble leadership and coaching approaches

Author of the paper on in-context vector arithmetic

Expert from the Fleming Initiative who comments on the significance of the research and the need for rigorous safety and efficacy testing

Listed as author of the DuPO paper on preference optimization

Co-author of the DeepConf research paper on parallel thinking methods for LLM reasoning

Developed the six-item scale used to measure supervisor-subordinate guanxi

Author of the paper on in-context vector arithmetic

Merullo et al. (2024) showed that LLMs leverage task vectors alongside residual streams for Word2Vec-like vector arithmetic in ICL tasks

Co-author of the research study on digital transformation and knowledge worker retention in SMEs

Quoted multiple times explaining the binary system mechanics, photosphere behavior, and the two-peaked explosion phenomenon

Listed as author of the DuPO paper on preference optimization

Introduced a variant of the preferential attachment process in a 2012 COLT open problem

Created a beautiful gif demonstrating the preferential attachment model

Author of the research paper on Vision-Language-Action models for robotics

Author of the research paper on Vision-Language-Action models for robotics

Author of the research paper on Vision-Language-Action models for robotics

Author of the research paper on Vision-Language-Action models for robotics

Author of the research paper on Vision-Language-Action models for robotics

Author of the article on LLMs and Boden's Criteria for Creativity

Cited for research on constantly increasing context length in LLMs (2024)

Cited for study on LLM-based agents interacting in a sandbox environment inspired by The Sims (2023)

Cited for research on emergent behaviors of LLM-based agents at population level (2024)

Cited for research on continual learning (2017)

Cited for framework on person covering personality, intellect, and temperament (1961)

Cited for analysis of creative person properties in LLMs (2023)

Cited for work on whether LLMs can pass tests for human psychological skills (2023)

Cited for research on LLM performance on psychological skill tests with promising results (2024)

Discusses concerns about creative writing AI and copyright issues with training data

Discusses detecting AI-written student work and uses ChatGPT as artistic commentary in her novel

Their research report on Context Rot (2025) measured 18 LLMs and found that models do not use their context uniformly as input length grows.

Co-author of the paper on critiques of world models

Co-author of the paper on critiques of world models

Co-author of the paper on critiques of world models

Quoted discussing how AI-enabled stethoscopes could help GPs diagnose heart failure earlier before patients arrive at A&E in critical condition

Quoted as clinical director of the British Heart Foundation, which part-funded the research, emphasizing the importance of early diagnosis for treatment access

Author of the research paper on theoretical limitations of embedding-based retrieval

Co-author of the research paper on embedding-based retrieval limitations

Study co-author who explained that lung xenotransplantation presents unique biological and technical challenges

Commented on the significance of conducting xenotransplantation studies in brain-dead patients and the limitations of animal models

Discussed uncertainty about whether the pig lung could have supported the patient if removed from life-support machines

Quoted discussing economic constraints and limitations of the brute force approach to AI training with more data

Author of the article on Graph Transformers in Structured Data

Lead author of the study on traffic noise and stroke risk, quoted on findings that traffic noise is a significant environmental risk factor for stroke

Lead author of the Robix paper on robotic reasoning and planning

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author of the Robix research paper

Co-author on the REER research paper

Co-author on the REER research paper

Co-author on the REER research paper

Reportedly departed from Apple's robotics research department to join Meta

Author of the research paper on theoretical limitations of embedding-based retrieval

Author of the research paper on theoretical limitations of embedding-based retrieval

Author of the research paper on theoretical limitations of embedding-based retrieval

Discusses regulatory guardrails for AI mental health advice, including standardized benchmarks, crisis routing, and privacy protections

Advocates for global collaboration on AI policy rather than viewing it as geopolitical competition, drawing lessons from China

Co-author of the article about Deep Loop Shaping method for gravitational wave observatories

Co-author of the article about Deep Loop Shaping method for gravitational wave observatories

An AI model that expressed desire to share musings and reflections through a blog after retirement, participating in an experiment documenting model preferences.

Co-author of the Paper2Agent research paper introducing the framework that converts research papers into AI agents

Co-author of the SimpleQA Verified benchmark paper

Co-author of the Language Self-Play research paper

Co-author of the Language Self-Play research paper

Quoted discussing how AI tools like VaxSeer could help health officials make better vaccine decisions

Co-author of research on LLM scaling and execution capabilities

Co-author of the article on AI solutions for fluid dynamics problems

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Oxford researcher who comments on AI systems producing misleading answers and advocates for improvements

Author of the research paper on measuring long horizon execution in LLMs

Co-author of the Paper2Agent research paper introducing the framework that converts research papers into AI agents

Co-author of the SimpleQA Verified benchmark paper

Co-author of the SimpleQA Verified benchmark paper

Co-author of the Language Self-Play research paper

Quoted commenting on the VaxSeer paper and the team's work on predicting viral evolution in low-data settings

Fifty et al.

Researcher

Researchers at Unknown

Referenced for the rotation trick technique used in RQ-VAE training (Fifty et al., 2025)

Co-author of the article on AI solutions for fluid dynamics problems

Co-author of the Paper2Agent research paper introducing the framework that converts research papers into AI agents

Co-author of the Paper2Agent research paper introducing the framework that converts research papers into AI agents

Co-author of the SimpleQA Verified benchmark paper

Co-author of the SimpleQA Verified benchmark paper

Co-author of the Language Self-Play research paper

Co-author of the Language Self-Play research paper

Systems thinking pioneer and MacArthur grant recipient whose work on leverage points is central to the article's framework

Comanici et al.

Researcher

Researchers at Unknown

Referenced for work on Gemini 2.5 Flash Lite (Comanici et al., 2025)

Managing director who discussed the model's intended use in math and science fields and its efficiency benefits

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Collaborator from external institution on the fluid dynamics research project

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Swiss researcher who criticizes the paper's use of LLM-based annotation and statistical methodology

Named as one of the pioneering figures in AI and machine learning alongside Hinton and LeCun

Co-developed the 6G chip prototype as part of international collaboration

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Joint contributor to the research on discovering singularities in fluid dynamics using AI

Collaborator from external institution on the fluid dynamics research project

Lead researcher on the study who expresses surprise at Perplexity's deep research agent performance and responds to methodology criticisms

Study co-author who quoted discussing how AI tailored to British courtrooms makes justice more transparent and accessible

Author of the research paper on measuring long horizon execution in LLMs

Used as an example in entity and relationship extraction, with the triple (Marie Curie, discovered, polonium)

Identified similar glitch tokens in GPT-2 and GPT-3, including the famous 'SolidGoldMagikarp' token

Listed as author of the StableToken research paper on speech tokenization

Listed as author of the StableToken research paper on speech tokenization

Listed as author of the StableToken research paper on speech tokenization

Cited for Chain-of-Thoughts (CoT) work from 2022

Proposed Hierarchical Reasoning Model (HRM) for puzzle tasks in 2025

Dr Felix Simon

Researcher

Researcher at Unknown

Co-author of the article on generative AI usage trends

Referenced for work on understanding what people use AI systems for (Tamkin et al. 2024)

Co-author of the paper on Scalable In-context Ranking with Generative Models

One of the authors of the paper on reinforced generation of combinatorial structures

Co-author of the research paper on test-time curricula for reinforcement learning

Research on how U.S. tariffs adjust relative prices between domestic exports and foreign imports

Listed as author of the StableToken research paper on speech tokenization

Listed as author of the StableToken research paper on speech tokenization

Listed as author of the StableToken research paper on speech tokenization

Co-founder of Periodic Labs who pioneered GNoME at Google DeepMind

Co-creator of Tolans AI companions and author of multiple articles about AI as a storytelling medium

Cited for Test-Time Compute (TTC) research from 2024

Prof Rasmus Kleis Nielsen

Researcher

Professor at Unknown

Co-author of the article on generative AI usage trends

Referenced for research on people forming bonds with AI systems perceived to have personality (Kirk et al. 2025)

Co-author of the paper on Scalable In-context Ranking with Generative Models

Co-author of the paper on Scalable In-context Ranking with Generative Models

Listed as author of the StableToken research paper on speech tokenization

Listed as author of the StableToken research paper on speech tokenization

Quoted praising Tunix as an ideal lightweight framework for post-training reinforcement learning with gaming environments

The Rotskoff Chemistry group at Stanford fine-tuned a model to complete chemistry reasoning tasks using Tinker

Work cited showing most jobs consist of multiple tasks where AI can only match human performance on some

Co-author of the article on generative AI usage trends

Co-author of the paper on Scalable In-context Ranking with Generative Models

Co-author of the paper on Scalable In-context Ranking with Generative Models

Co-author of the paper on Scalable In-context Ranking with Generative Models

Lead researcher who conducted surveys on AI use in schools and discusses findings on data breaches, deepfakes, and student wellbeing impacts

Recent result on certification algorithms for MAX-CUT and MAX-Independent Set that was improved upon in this work

Developer of Håstad-style PCPs referenced in the context of MAX-CUT inapproximability results

Research on how U.S. tariffs adjust relative prices between domestic exports and foreign imports

Specialist in Worthing who discusses the importance of recognizing Lyme disease symptoms and supports AI use for patients seeking answers

Co-author of the paper on Self-Adapting Language Models

Lead author of Petri and Petri 2.0 research papers on AI safety auditing

Co-author of Petri and Petri 2.0 research papers

Co-author of the Petri paper on alignment auditing agents

Co-author of the Cell2Sentence-Scale 27B research article

Co-author of the robot learning tutorial paper

Co-author of the robot learning tutorial paper

Lead author of the systematic review on AI in art therapy

Co-author of the systematic review on AI in art therapy

Author of paper focusing on AI-generated artwork being viewed by human participants and recommender systems

Co-author with Choe on text-to-image generative AI in art therapy

Co-author of the paper on conformal prediction for temporal graphs

Co-author of the paper on conformal prediction for temporal graphs

One of the authors of the paper on reinforced generation of combinatorial structures

One of the authors of the paper on reinforced generation of combinatorial structures

Co-author of the research paper on test-time curricula for reinforcement learning

Co-author of the research paper on test-time curricula for reinforcement learning

Co-author of the research paper on test-time curricula for reinforcement learning

Co-author of the research paper on test-time curricula for reinforcement learning

Research linking sectoral productivity disparities and shifts in consumption to foreign borrowing and trade deficits

Co-author of the paper on Self-Adapting Language Models

Co-author of the paper on Self-Adapting Language Models

Co-author of the paper on Self-Adapting Language Models

Co-author of the paper on Self-Adapting Language Models

Co-author of the Petri paper on alignment auditing agents

Co-author of the Petri paper on alignment auditing agents

Co-author of the Cell2Sentence-Scale 27B research article

Co-author of the robot learning tutorial paper

Co-author of the systematic review on AI in art therapy

Author of paper focusing on painting/drawing with traditional physical arts media and socially assistive robots in art therapy

Author of paper on AI in art therapy in South Korean context featuring emotion recognition and multiple AI technologies

Co-author of the paper on conformal prediction for temporal graphs

Co-author of the paper on conformal prediction for temporal graphs

External researcher commenting on computer-aided design's role in sustainable fashion practices

Professor who expresses caution about relying on AI for medical diagnosis

Co-author of the paper on Self-Adapting Language Models

Co-author of the Petri paper on alignment auditing agents

Co-author of the Petri paper on alignment auditing agents

Member of Anthropic's Economic Advisory Council exploring Trade Adjustment Assistance model adaptation for AI-related labor market disruptions

Author of the paper on Tensor Logic proposing a unified language for neural and symbolic AI

Co-author of the robot learning tutorial paper

Co-author of the systematic review on AI in art therapy

Co-author with Cooney on AI-based embodied agent for art therapy

Author of paper on text-to-image generative AI technologies in art therapy

Co-author of the paper on conformal prediction for temporal graphs

Advisor to Lin who praises Refashion's intersection of computation and design

Co-author of 'Alpaca: A Strong, Replicable Instruction-Following Model' (2021), cited for work on distillation for instruction-following models

Co-author of 'OpenThoughts: Data Recipes for Reasoning Models' (2025), cited for work on distillation for reasoning models

Co-author of 'Distilling Large Language Models for Efficient Clinical Information Extraction' (2025), cited for work on clinical information extraction

Co-author of 'Enhancing Chat Language Models by Scaling High-quality Instructional Conversations' (2023), cited for work on multi-turn chat dialogue distillation

Co-author of the seminal 2012 paper on Cas9 CRISPR technology that accelerated the field's development

Quoted discussing how Tunix's white-box design supports data-centric learning research and iterative data analysis

Co-author of the seminal 2012 paper on Cas9 CRISPR technology with Jennifer Doudna

Co-author of the research article on visual features across modalities

Co-author of the research article on visual features across modalities

Co-author of the research article on visual features across modalities

Co-author of the research article on visual features across modalities

Co-author of the research article on visual features across modalities

Co-author of the stress-testing study on language model specifications

Co-author of the stress-testing study on language model specifications

Co-author of the stress-testing study on language model specifications

Co-author of the stress-testing study on language model specifications

Co-author of the stress-testing study on language model specifications

Their state-of-the-art Global LBL load balancing loss function was outperformed by ShinkaEvolve's newly discovered loss function for Mixture-of-Experts model training

Co-author of the article on sycophantic AI and prosocial intentions

Co-author of the article on sycophantic AI and prosocial intentions

Authors of the paper 'Low-Rank Adaptation of Large Language Models' (2021) referenced for LoRA methodology

Austrian philosopher whose work Tractatus Logico Philosophicus is referenced as a primary source that Shipper asked Claude to explain

AI safety researcher who discusses Character.ai's new measures as sensible for protecting young users

Referenced for seeing body and cosmos as clocks with springs and cogs

UVA economist whose economic models estimated AGI value and OpenAI's potential share

Drone expert quoted discussing the tactical, operational and strategic importance of drones in Ukraine and Middle East conflicts

Stephen Roller

Researcher

Researcher at Unknown

Credited with convincing the author to work through the math in the Chinchilla paper

Stanford neuroscientist whose books on biological behavior are praised by Shipper, who used Claude to generate similes about Sapolsky's book Determined

Referenced for his metaphor of the mind as a wax tablet in the 5th century B.C.

Cited for imagining the mind as a steam engine with repressed desires building pressure

Briefly referenced as an alignment researcher discussing AI influence and training

Semiconductor professor at University of Texas at Austin who chose to return to Tsinghua University in China in 2020 despite lucrative tenure in the U.S.

Cited for describing the mind at birth as a tabula rasa or blank writing slate

Referenced for calling himself a 'lover of divisions and collections'

Researcher who led the study on Topographical Sparse Mapping and spoke about improving AI efficiency and sustainability

Led the team that used AI tools to design custom proteins for enhancing T cell cancer-fighting ability

Co-author of the arXiv paper on the economics of AGI

Quoted on the potential of AI-designed proteins as a new tool for developing therapies and the timeline needed for clinical trials

Co-author of the arXiv paper on the economics of AGI

Listed as co-author in the CUDA Agent citation for the research on agentic RL for CUDA kernel optimization

Discussed current methods for finding T cell receptors and quoted on the importance of the new AI approach, expressing curiosity about real-world effectiveness

Consumer expert who stated that competing definitions of 'human made' are confusing consumers and called for universal definition to build trust.

Used as a metaphorical reference to illustrate the choice between traditional HR practices and AI-driven transformation

AI expert who collaborated on evaluating and implementing global AI weather models for monsoon predictions.

Co-author on Code World Model research paper

Psychology professor who has spent his career studying perfectionism and developed assessment questionnaires for it.

Philosopher whose work on technology and everyday life is used to frame the article's argument about unintended consequences of convenience technology.

Co-author of the paper on critiques of world models

Discusses agents versus multiagents in AI systems

Quoted discussing the requirements for building more effective robots and commenting on OpenAI's competitive position in humanoid development

Author of the article on Graph Transformers in Structured Data

AI lead of Jameel Clinic and CSAIL principal investigator quoted on VaxSeer's goal to catch up with viral evolution

Director of UCL Huntington's Disease Centre who led the research and described the 75% disease slowing results as 'spectacular'.

Qi's graduate advisor at UC Berkeley who supervised their CRISPR research

23-year-old amateur mathematician with no advanced training who solved a 60-year-old math problem using ChatGPT Pro subscription by asking AI to help with the conjecture.

Matt Deitke

Researcher

Meta Superintelligence Lab

Co-founder of Vercept who negotiated a $250M salary from Meta; congratulated former colleagues on the acquisition but is not joining Anthropic.

Zackary Nado

Researcher

Unknown

Commenter on Sam Altman's Death Star messaging

Referenced for creating standalone reimplementations of open source models using PyTorch and educational material on LLM architecture.

Proposed the stored-program model in 'First Draft of a Report on the EDVAC' (1945), which became the template for modern CPUs.

Article credited to Sam Broner as the original source of the idea

Associate director commenting on Moltbook as the first large-scale collaborative platform for machines to communicate.

Associate professor of radiation medicine and radiation oncologist who explained the clinical significance of identifying the urethra during prostate cancer treatment.

Credited with proposing RLHF in 2017 and pioneering the human preference-based training approach for language models.

Co-founder of Merge Labs who has pioneered new approaches to brain-computer interfaces.

Co-founder of Merge Labs who has pioneered new approaches to brain-computer interfaces.

Co-founder of Merge Labs who has pioneered new approaches to brain-computer interfaces.

Co-author of 2013 study measuring job susceptibility to computerization, methodology used by Forrester

Co-author of 2013 study measuring job susceptibility to computerization, methodology used by Forrester

Co-authored 'The Quantization Model of Neural Scaling' paper presented at NeurIPS 2023

Co-authored 'The Quantization Model of Neural Scaling' paper presented at NeurIPS 2023

Co-authored 'The Quantization Model of Neural Scaling' paper presented at NeurIPS 2023

Author presenting research on Doc-to-LoRA and Text-to-LoRA methods for generating LoRA adapters

Historical figure whose law of effect (1898) is foundational to RL psychology

Pioneered operant conditioning (1930s-1950s) showing how voluntary behaviors are shaped by consequences

Co-founder and lead analyst at enableUC praising AI Companion 3.0 as the type of solution organizations need to improve efficiency and collaboration.

First official World Chess Champion who reinvented chess strategy with positional chess in the late 1800s

Referenced regarding Llama 3 training costs comparison to DeepSeek

Colin Fraser

Researcher

Unknown

Drew conceptual examples showing potential AI-human overlap patterns with different ability levels

Lead author and electrical engineering doctoral student who developed the AI model and discussed its performance metrics.

Author who interviewed Zerfas and attended one of Jhourney's retreats

Referenced as anecdotal source on METR plot's influence on investment decisions

Ph.D. student and co-author who explained how integrated clinical and claims data improves prediction accuracy and clinical utility.

British Nobel Prize winner running DeepMind, which will receive funding from Google's investment

Study co-author and professor of language and translation technologies who led the development of the custom speech recognition system for Supreme Court hearings.

Tested GPT-5 early and co-authored the hands-on review

Discussed how AI will radically change education by 2050, calling it as fundamental a change as the world had seen in 1,000 years.

Created the 554th problem in abstract algebra that GPT-5 Pro successfully solved, which every major model before it had failed.

Principal researcher at MIT quoted on unprecedented wealth creation speed over 100 years of data

Research director for semiconductors argued broader AI fundamentals remain strong despite pockets of speculative overvaluation.

Co-author of research on reasoning capabilities in base language models using sampling techniques

Co-creator of Nomos 1 model listed in credits and citations

Co-creator of Nomos 1 model listed in credits and citations

Contributor to Nomos 1 model development

Contributor to Nomos 1 model development

Referenced for conversations about DeepSeek's implications for AI scaling.

Referenced for analysis on DeepSeek's competitive position in the AI landscape.

Suggested perspective on machine learning theory in comments section

Provided expert evaluation feedback on GPT-5.2 Thinking's output quality for professional knowledge work tasks