A Definition of AGI

Paper

Dan Hendrycks¹, Dawn Song², Christian Szegedy³, Honglak Lee⁴, Yarin Gal⁵

Erik Brynjolfsson⁶, Sharon Li⁷, Andy Zou^1,8,9, Lionel Levine¹⁰, Bo Han¹¹, Jie Fu¹², Ziwei Liu¹³

Jinwoo Shin¹⁴, Kimin Lee¹⁴, Mantas Mazeika¹, Long Phan¹, George Ingebretsen¹

Adam Khoja¹, Cihang Xie¹⁵, Olawale Salaudeen¹⁶, Matthias Hein¹⁷, Kevin Zhao¹⁸

Alexander Pan², David Duvenaud^19,20, Bo Li²¹, Steve Omohundro²², Gabriel Alfour²³

Max Tegmark¹⁶, Kevin McGrew²⁴, Gary Marcus²⁵, Jaan Tallinn²⁶

Eric Schmidt¹⁶, Yoshua Bengio^27,28

Dan Hendrycks¹, Dawn Song², Christian Szegedy³

Honglak Lee⁴, Yarin Gal⁵, Erik Brynjolfsson⁶

Sharon Li⁷, Andy Zou^1,8,9, Lionel Levine¹⁰

Bo Han¹¹, Jie Fu¹², Ziwei Liu¹³

Jinwoo Shin¹⁴, Kimin Lee¹⁴, Mantas Mazeika¹

Long Phan¹, George Ingebretsen¹, Adam Khoja¹

Cihang Xie¹⁵, Olawale Salaudeen¹⁶, Matthias Hein¹⁷

Kevin Zhao¹⁸, Alexander Pan², David Duvenaud^19,20

Bo Li²¹, Steve Omohundro²², Gabriel Alfour²³

Max Tegmark¹⁶, Kevin McGrew²⁴, Gary Marcus²⁵

Jaan Tallinn²⁶, Eric Schmidt¹⁶, Yoshua Bengio^27,28

¹Center for AI Safety

²University of California, Berkeley

³Morph Labs

⁴University of Michigan

⁵University of Oxford

⁶Stanford University

⁷University of Wisconsin–Madison

⁸Gray Swan AI

⁹Carnegie Mellon University

¹⁰Cornell University

¹¹Hong Kong Baptist University

¹²HKUST

¹³Nanyang Technological University

¹⁴KAIST

¹⁵University of California, Santa Cruz

¹⁶Massachusetts Institute of Technology

¹⁷University of Tübingen

¹⁸University of Washington

¹⁹University of Toronto

²⁰Vector Institute

²¹University of Chicago

²²Beneficial AI Research

²³Conjecture

²⁴Institute for Applied Psychometrics

²⁵New York University

²⁶CSER

²⁷Université de Montréal

²⁸LawZero

Introduction

The lack of a concrete definition for Artificial General Intelligence (AGI) obscures the gap between today’s specialized AI and human-level cognition. This paper introduces a quantifiable framework to address this, defining AGI as matching the cognitive versatility and proficiency of a well-educated adult. To operationalize this, we ground our methodology in Cattell-Horn-Carroll theory, the most empirically validated model of human cognition.

The framework dissects general intelligence into ten core cognitive domains—including reasoning, memory, and perception—and adapts established human psychometric batteries to evaluate AI systems. Application of this framework reveals a highly “jagged” cognitive profile in contemporary models. While proficient in knowledge-intensive domains, current AI systems have critical deficits in foundational cognitive machinery, particularly long-term memory storage.

The resulting AGI scores (e.g., GPT-4 at 27%, GPT-5 at 58%) concretely quantify both rapid progress and the substantial gap remaining before AGI.

GPT-4 and GPT-5 capabilities radar chart

The capabilities of GPT-4 and GPT-5.

Definition

"AGI is an AI that can match or exceed the cognitive versatility and proficiency of a well-educated adult."

The framework comprises ten core cognitive components, derived from CHC broad abilities and weighted equally (10%) to prioritize breadth and cover major areas of cognition:

Acquired Knowledge

Perception

Central Executive

Output

Citation

@misc{hendrycks2025agidefinition,
      title={AGI Definition}, 
      author={Dan Hendrycks and Dawn Song and Christian Szegedy and Honglak Lee and Yarin Gal and Erik Brynjolfsson and Sharon Li and Andy Zou and Lionel Levine and Bo Han and Jie Fu and Ziwei Liu and Jinwoo Shin and Kimin Lee and Mantas Mazeika and Long Phan and George Ingebretsen and Adam Khoja and Cihang Xie and Olawale Salaudeen and Matthias Hein and Kevin Zhao and Alexander Pan and David Duvenaud and Bo Li and Steve Omohundro and Gabriel Alfour and Max Tegmark and Kevin McGrew and Gary Marcus and Jaan Tallinn and Eric Schmidt and Yoshua Bengio},
      year={2025},
}