News

Ending global plastic pollution may sound like a daunting task, but that’s exactly what the United Nations set out to do in 2022 by creating the first-ever international legally binding treaty on plastic pollution. DSE contributed to ongoing negotiations by designing an open source, AI-powered model that enables policymakers to examine combinations of policies at different levels to inform their negotiation positions. Relatedly, we published new findings in Science in 2024 and 2026 that recommend a policy package to reduce global plastic pollution by 91% by 2050 and strategies to improve data...

On this week's episode of Boiling Point, postdoctoral researcher Cassie Buhler discusses the benefits (particularly for energy efficiency and privacy) of open weight AI models.

In this recent article, faculty advisor Carl Boettiger describes how he teaches his students to critically assess any AI-based findings, rather than accept them as objective facts.

Our perspectives on the value of open and local AI models in science

Health Medicine Net's recent story on DSE's commentary in Nature and interview with Berkeley News.

GeoJupyter's flagship tool to enhance open collaboration on geospatial data analysis is published for the first time in an academic journal.

A reshare of Packaging Insights' story on our recent Science publication that proposes a global reporting framework within the United Nations Global Plastics Treaty.

A feature on faculty co-director Doug McCauley's recent paper in Science regarding the United Nations' ongoing Global Plastics Treaty negotiations.