Dan Hendrycks (born ) is an American

machine learning Machine learning (ML) is a field of study in artificial intelligence concerned with the development and study of Computational statistics, statistical algorithms that can learn from data and generalise to unseen data, and thus perform Task ( ...

researcher. He serves as the director of the

Center for AI Safety The Center for AI Safety (CAIS) is a nonprofit organization based in San Francisco, that promotes the safe development and deployment of artificial intelligence (AI). CAIS's work encompasses research in technical AI safety and AI ethics, advoca ...

, a nonprofit organization based in

San Francisco San Francisco, officially the City and County of San Francisco, is a commercial, Financial District, San Francisco, financial, and Culture of San Francisco, cultural center of Northern California. With a population of 827,526 residents as of ...

California California () is a U.S. state, state in the Western United States that lies on the West Coast of the United States, Pacific Coast. It borders Oregon to the north, Nevada and Arizona to the east, and shares Mexico–United States border, an ...

Early life and education

Hendrycks was raised in a Christian

evangelical Evangelicalism (), also called evangelical Christianity or evangelical Protestantism, is a worldwide, interdenominational movement within Protestantism, Protestant Christianity that emphasizes evangelism, or the preaching and spreading of th ...

household in

Marshfield, Missouri Marshfield is a city in and the county seat of Webster County, Missouri. As of the 2020 census, Marshfield had a population of 7,458. As of 2024, the population is estimated to be 7,935. It is part of the Springfield, Missouri, metropolitan ar ...

. He received a B.S. from the

University of Chicago The University of Chicago (UChicago, Chicago, or UChi) is a Private university, private research university in Chicago, Illinois, United States. Its main campus is in the Hyde Park, Chicago, Hyde Park neighborhood on Chicago's South Side, Chic ...

in 2018 and a

Ph.D. A Doctor of Philosophy (PhD, DPhil; or ) is a terminal degree that usually denotes the highest level of academic achievement in a given discipline and is awarded following a course of graduate study and original research. The name of the deg ...

from the

University of California, Berkeley The University of California, Berkeley (UC Berkeley, Berkeley, Cal, or California), is a Public university, public Land-grant university, land-grant research university in Berkeley, California, United States. Founded in 1868 and named after t ...

Computer Science Computer science is the study of computation, information, and automation. Computer science spans Theoretical computer science, theoretical disciplines (such as algorithms, theory of computation, and information theory) to Applied science, ...

in 2022.

Career and research

Hendrycks' research focuses on topics that include machine learning safety,

machine ethics Machine ethics (or machine morality, computational morality, or computational ethics) is a part of the ethics of artificial intelligence concerned with adding or ensuring moral behaviors of man-made machines that use artificial intelligence, otherw ...

, and robustness. He credits his participation in the

effective altruism Effective altruism (EA) is a 21st-century philosophical and social movement that advocates impartially calculating benefits and prioritizing causes to provide the greatest good. It is motivated by "using evidence and reason to figure out how to b ...

(EA) movement-linked

80,000 Hours 80,000 Hours is a London-based nonprofit organisation that conducts research on which careers have the largest positive social impact and provides career advice based on that research. The organisation's name refers to the typical amount of tim ...

program for his career focus towards AI safety, though denied being an advocate for EA. Hendrycks is the main author of the research paper that introduced the

activation function The activation function of a node in an artificial neural network is a function that calculates the output of the node based on its individual inputs and their weights. Nontrivial problems can be solved using only a few nodes if the activation f ...

GELU in 2016, and of the paper that introduced the language model benchmark MMLU (Massive Multitask Language Understanding) in 2020. In February 2022, Hendrycks co-authored recommendations for the US

National Institute of Standards and Technology The National Institute of Standards and Technology (NIST) is an agency of the United States Department of Commerce whose mission is to promote American innovation and industrial competitiveness. NIST's activities are organized into Outline of p ...

(NIST) to inform the management of risks from

artificial intelligence Artificial intelligence (AI) is the capability of computer, computational systems to perform tasks typically associated with human intelligence, such as learning, reasoning, problem-solving, perception, and decision-making. It is a field of re ...

. In September 2022, Hendrycks wrote a paper providing a framework for analyzing the impact of AI research on societal risks. He later published a paper in March 2023 examining how

natural selection Natural selection is the differential survival and reproduction of individuals due to differences in phenotype. It is a key mechanism of evolution, the change in the Heredity, heritable traits characteristic of a population over generation ...

and competitive pressures could shape the goals of

artificial agents In artificial intelligence, an intelligent agent is an entity that perceives its environment, takes actions autonomously to achieve goals, and may improve its performance through machine learning or by acquiring knowledge. Leading AI textbook ...

. This was followed by "An Overview of Catastrophic AI Risks", which discusses four categories of risks: malicious use, AI race dynamics, organizational risks, and rogue AI agents. Hendrycks is the safety adviser of xAI, an AI startup company founded by

Elon Musk Elon Reeve Musk ( ; born June 28, 1971) is a businessman. He is known for his leadership of Tesla, SpaceX, X (formerly Twitter), and the Department of Government Efficiency (DOGE). Musk has been considered the wealthiest person in th ...

in 2023. To avoid any potential conflicts of interest, he receives a symbolic

one-dollar salary A one-dollar salary is a method used when a business executive or government employee wishes to work without direct compensation, but must receive a salary above zero to legally distinguish them from a volunteer. The concept first emerged in the ...

and holds no company equity. In November 2024, he also joined

Scale AI Scale AI, Inc. is an American data annotation company based in San Francisco, California. It provides data labeling and model evaluation services to develop applications for artificial intelligence. The company’s research arm, the Safety, E ...

as an advisor collecting a one-dollar salary. Hendrycks is the creator of

Humanity's Last Exam Humanity's Last Exam (HLE) is a language model benchmark consisting of 2,500 questions across a broad range of subjects. It was created jointly by the Center for AI Safety and Scale AI. Creation Stanford HAI's AI Index 2025 Annual Report cites ...

, a benchmark for evaluating the capabilities of

large language model A large language model (LLM) is a language model trained with self-supervised machine learning on a vast amount of text, designed for natural language processing tasks, especially language generation. The largest and most capable LLMs are g ...

s, which he developed in collaboration with Scale AI. In 2024 Hendrycks published a 568 page book entitled "Introduction to AI Safety, Ethics, and Society" based on courseware he had previously developed.

Selected publications

* * * *

References

{{DEFAULTSORT:Hendrycks, Dan Machine learning researchers Living people University of California, Berkeley alumni American computer scientists 1990s births People from Marshfield, Missouri AI safety scientists