Skip to content
Vol. 1Global Edition
Icon MagazineProfiles of the people who build
Data: TradingView

From Quantum Physics to AI Safety: Jared Kaplan's Guiding Hand at Anthropic

Jared KaplanCo-founder and Chief Science Officer, Anthropic12 November 20262 min read

Jared Kaplan, a distinguished theoretical physicist, has carved a unique path from the intricate realms of quantum gravity to the forefront of artificial intelligence development. As a co-founder and Chief Science Officer at Anthropic, he brings a rigorous scientific approach to shaping advanced AI systems, with a particular focus on their safe and responsible deployment.

Academic Foundations

Kaplan’s intellectual journey began with a strong foundation in mathematics and physics. He attended the Illinois Mathematics and Science Academy before pursuing a bachelor’s degree in physics and mathematics at Stanford University. His academic pursuits continued at Harvard University, where he earned a PhD in physics. His doctoral research explored aspects of holography, guided by a prominent figure in theoretical physics. Following his doctoral studies, Kaplan engaged in postdoctoral research at SLAC and Stanford University, further deepening his expertise. Since 2012, he has served as a professor of physics at Johns Hopkins University, where his research interests span complex areas such as quantum gravity, holography, conformal field theory, and related topics within particle physics and cosmology. This extensive background in fundamental science provided him with a robust analytical framework, which would later prove instrumental in his contributions to the field of artificial intelligence. His academic contributions have been recognized through various accolades, including a Hertz Fellowship, a Sloan Research Fellowship, and an NSF CAREER award.

Transition to Machine Learning

In 2019, Kaplan transitioned his formidable analytical skills to the burgeoning field of machine learning, joining OpenAI as a researcher. This move marked a pivotal shift in his career, allowing him to apply his scientific rigor to the challenges of artificial intelligence. During his tenure, he made significant contributions to foundational research in large language models. He co-authored the influential 2020 paper, Scaling Laws for Neural Language Models, which empirically demonstrated that the performance of language models consistently improves with increases in their size, the volume of training data, and the computational resources utilized. This work provided crucial insights into the predictable behavior of large-scale AI systems. In the same year, he also co-authored Language Models are Few-Shot Learners, a paper that introduced GPT-3, a significant advancement in the capabilities of language models. Additionally, Kaplan was involved in the development of Codex, further solidifying his impact on the practical application and advancement of AI technologies. His work at OpenAI established him as a key contributor to the understanding and development of modern AI.

Guiding Anthropic’s Scientific and Safety Trajectory

Building on his extensive experience, Kaplan co-founded Anthropic, an organization established with the explicit goal of promoting AI safety. As Anthropic’s Chief Science Officer, he plays a central role in guiding the company’s research and development efforts. His leadership ensures that scientific inquiry underpins the creation of Anthropic’s flagship products, such as the Claude series of large language models. In a further commitment to responsible AI development, Anthropic announced in October 2024 that Kaplan would also assume the role of Responsible Scaling Officer. In this capacity, he is entrusted with the critical responsibility of defining the safety assessments and necessary precautions that must be implemented before any new model is released. This role underscores his dedication to ensuring that advancements in AI are balanced with thorough consideration for their potential implications, making him a key figure in shaping the ethical and scientific trajectory of advanced AI systems.

Sources

  1. Wikipedia: Jared Kaplan
  2. Wikipedia: Anthropic
  3. Wikidata

Written by Icon Magazine from these sources, and fact-checked against them.

Keep reading

More profiles