When India crashes into Eurasia: faults, rising mountains and giant earthquakes

Jean-Philippe Avouac, Earle C. Anthony Professor of Geology, California Institute of Technology

Plate tectonics has revealed how India split away from Africa and crushed into Eurasia. With geodesy, we can see the process going on, driving widespread continental deformation, mountain building and large earthquakes. The Himalaya stands out for its majestic size, rapid deformation rates and the seismic hazard it poses to the dense population of northern India and surrounding areas. Earthquakes occur continuously as India is thrust under the Himalaya and Tibet and is rocked by the monsoon. Some of them must occasionally grow to very large magnitudes. While current technology cannot accurately predict the exact timing of a major earthquake, the probability of the location, and occurrence time and magnitude of future large events can be assessed. This knowledge is crucial for implementing effective mitigation strategies, improving building codes, and enhancing public preparedness. Elected in 2025, Avouac represents Class I’s section on Physical and Mathematical Sciences.

The Cellular Biology of Words

Edward F. Chang, Professor and Chair, Neurological Surgery, University of California, San Francisco

How do words arise from the activity of individual human neurons? In this talk, I will describe recent work using Neuropixel recordings in humans to probe speech and language at cellular resolution. By sampling thousands of neurons simultaneously during speech perception and production, we reveal how words are encoded as structured population codes, how these codes are transformed through core cortical computations, and how the same neural populations participate in both perception and action. These results move beyond correlates to define elements of the neural code for speech, linking single-cell activity to phonetic features, word structure, and generative processes. Together, this work reframes speech and language as a problem in human cellular biology and establishes a foundation for mechanistic models—and future clinical restoration—of communication. Elected in 2025, Chang represents Class II’s section on Biological Sciences.

Computational Foundations of Modern Robotics

Lydia E. Kavraki, Kenneth and Audrey Kennedy Professor of Computing, Rice University

Over the past sixty years, robots have progressed from isolated industrial artifacts to interactive, semi-autonomous agents increasingly embedded in next-generation factories, hospitals, and homes. Yet, realizing robots that collaborate safely, robustly, and transparently with people and with one another continues to pose deep theoretical and practical challenges. This talk will trace the evolution of the sampling-based paradigm in robot motion planning and highlight its central role in enabling reasoning under uncertainty, long-horizon autonomy, and human-centered collaboration. It will discuss how motion planning confronts the fundamental tension between the complexity of high-dimensional robotic systems and the stringent constraints imposed by physical embodiment, including dynamics, sensing, and real-world interaction. Building on this foundation, the talk will explore the relationship between motion planning and higher-level decision-making, where specifications can define what a robot must accomplish rather than how it should accomplish it, and the frameworks required to support more capable and reliable robots that serve people in shared environments. Elected in 2025, Kavraki represents Class III’s section on Engineering and Applied Sciences.

Humanizing Drug Discovery 

David Matthew Altshuler, Executive Vice President and Chief Scientific Officer, Vertex Pharmaceuticals, Inc.

Despite great progress in medical science much of human biology is not yet understood, and existing disease therapies are of limited benefit. These concepts are inexorably linked, as causal human disease biology is fundamental to discover and develop disease modifying therapy and eventual cures. Over the past 40 years human genetics has enabled connection of clinical phenotypes to molecular disease mechanisms, first for rare single gene disorders and more recently for common complex diseases.  These discoveries enable (a) generation and testing of mechanistic hypotheses, (b) improved prediction of disease risk and/or treatment response, and (c) identification of therapeutic targets.  Until recently, many such potential therapeutic targets were not “druggable” using existing technologies. I will discuss development of human genetics as a method to study human disease, to guide target selection for therapeutic discovery, and how advances in therapeutic technologies have expanded the ability to address causal factors in human disease. Elected in 2025, Altshuler represents Class VI’s section on Biomedical Sciences.

Thinking Different and Collective Intelligence

Scott E. Page, John Seely Brown Distinguished University Professor, University of Michigan

Human collective intelligence takes two forms. Additive collective intelligence arises when a team consistently outperform their best member at a given task exhibit Emergent collective intelligence arises when a collective possess a functionality not present in its members. Both forms require cognitive diversity. Diversity, though, is not sufficient. While arbitrary cogntive difference can produce better predictions, solutions, and ideas, the most effective groups leverage diversity that has been cultivated or had time to evolve within thoughtfully designed protocols and institutions. Elected in 2025, Page represents Class V’s section on Behavioral and Social Sciences.

The Hidden Language of Plants: How Genome Evolution Drives Chemical Diversity

C. Robin Buell, Georgia Research Alliance Eminent Scholar Chair in Crop Genomics, University of Georgia

Plants and their natural products have been harnessed by humans for millennia for their health promoting activities, insecticidal and anti-microbial activities, as well as fragrance and food additive properties. These natural products, or specialized metabolites, have diverse functions in plants as pollinator attractants, defense and signaling molecules, and components of cell walls. The sheer diversity of specialized metabolites in plants is attributable to the dynamics of gene and genome evolution which are catalysts of species diversification. The advent of genomics technologies has revolutionized the discovery of the genes encoding specialized metabolites revealing a myriad of genome features and evolutionary mechanisms that underlie extant chemodiversity. In my talk, I will highlight the vinca alkaloids of Madagascar periwinkle, currently used in the clinic as anti-cancer agents, and how the biosynthetic pathway has diverged resulting in alkaloid diversity between accessions, organs, and cell types. Elected in 2025, Buell represents Class VI’s section on Applied Biological, Agricultural, and Environmental Sciences.