From Greenland to the Seafloor, Columbia Scientists Put AI to Work
A Columbia team has trained AI to approximate in three seconds a Greenland ice-sheet simulation that normally takes a week to run, allowing scientists to test far more scenarios to better understand how quickly the ice will melt and how much it could raise sea levels along the world’s coastlines.
Math professors are reconsidering their pedagogy as AI models solve the world’s biggest—and smallest—math problems.
The world’s power grids deserve credit as arguably the largest, most complex machines humans have ever built. They also deserve some sorely needed attention: they’re currently under more pressure than ever before.
Columbia researchers found that AI can analyze teens’ smartphone text to predict depressive symptoms and mood changes. They aim to develop an app that offers timely mental health support through personalized activities and problem-solving techniques.
Columbia has banned digital devices in two Core Curriculum courses to reduce distractions and encourage critical thinking, deeper reading, and in-person discussion.
Columbia Applied Analytics alumnus Zain Merchant shares how his degree, particularly its natural language analytics coursework, prepared him for the AI boom and helped him build a career in data science, emphasizing adaptability, critical thinking, and communicating insights effectively.
The first conversation, titled “AI + Human Development” and moderated by CNN host Fareed Zakaria, featured Greek Prime Minister Kyriakos Mitsotakis and World Bank Group President Ajay Banga
Should companies test products in their familiar, local pond or in the new market where they want to scale?
Michael Morris, Chavkin-Chang Professor of Leadership, Management Division is the author of this Fortune commentary, “I study tribal psychology and build AI agents for my business students—the rogue OpenAI ‘swarm’ alarmed me”.
Stijn Van Nieuwerburgh featured in The Wall Street Journal on the economics of the AI investment boom. His research is cited as the basis for an estimate that AI companies would need to generate roughly $3.5 trillion in annual revenue by 2032—about 8.8% of U.S. GDP—to justify the enormous capital being invested in AI infrastructure.
Columbia and City College researchers are developing an AI-powered yeast sensor that can detect PFAS forever chemicals in drinking water faster and more cheaply than conventional testing methods.
Columbia’s School of the Arts and SEAS are launching an AI and Arts, Creativity, and Media concentration that combines artificial intelligence with creative fields such as storytelling, film, photography, and digital media.
For a data center investor, a signed lease with a major technology company is a powerful selling point. The tenant has deep pockets, demand for computing capacity is growing, and access to power limits competing supply. Put those features together, and a new campus can look like an unusually attractive real estate investment.
As the government prioritizes AI development and national security over environmental concerns, the xAI lawsuit raises questions about corporate accountability, environmental justice, and the future of citizen-led enforcement.
A monthly newspaper, distributed as a PDF, is trying to imagine a future for journalism beyond the reach of Silicon Valley.
Big Tech has spent years resisting accountability. As the federal government adopts its own evasion tactics—locking out the press, and now investigating news networks—it’s worth remembering journalism’s essential democratic function.
Growth driven by date center development comes with hidden economic risks, according to a new paper.
Education technology founder Luyen Chou’s (M.A. ’07) newest organization is using AI to support learning in classrooms and beyond
Launched this fall, the AI Initiative reflects a new approach to preparing lawyers for a profession increasingly shaped by AI.
Buckmaster walked a packed room through his AI-assisted fluid proof and a faculty panel ruminated over AI’s role in math.
A Columbia team has trained AI to approximate in three seconds a Greenland ice-sheet simulation that normally takes a week to run, allowing scientists to test far more scenarios to better understand how quickly the ice will melt and how much it could raise sea levels along the world’s coastlines.