Jared Kaplan

Jared Kaplan is a theoretical physicist and AI researcher who co-founded Anthropic. He studied physics and mathematics at Stanford University and earned his doctorate in physics at Harvard University. His academic career included work at Johns Hopkins University on theoretical questions before his research expanded into the behavior of large neural models.

Kaplan was a lead author of Scaling Laws for Neural Language Models in 2020. The research examined relationships among model size, data, computing resources, and predictive performance. These empirical relationships helped researchers reason about resource allocation, while remaining dependent on the settings and assumptions under which the experiments were conducted.

He also contributed to the GPT-3 research published in 2020 and became one of Anthropic's co-founders in 2021. His move into AI connected a quantitative research background with experiments on large language models. At Anthropic, this work developed alongside broader efforts to understand model capabilities, behavior, and safety.

Anthropic identifies Kaplan as its chief science officer, responsible for foundational technical research and the development of safer systems. He was also named responsible scaling officer under the company's updated policy in 2024. His contribution combines empirical investigation of model performance with responsibilities for how capability growth is assessed and connected to decisions about development and deployment.