UT Austin Brings Together 600 AI Leaders to Address Workforce, Ethics Questions
More than 600 researchers, industry executives, and government officials gathered at the University of Texas at Austin this month to examine how artificial intelligence, machine learning, and robotics are reshaping work, healthcare, and defense. The Texas Symposium on Machine Learning, Responsible AI, and Robotics tackled concrete problems: how agentic AI will change employment, whether AI companions pose psychological risks, and what safeguards protect human agency as autonomous systems proliferate.
The inaugural two-day event, organized by Texas Robotics, the Machine Learning Lab, and Good Systems - Ethical AI at UT Austin, marked the first time these three research groups held a joint symposium. The programming mixed panels, research presentations, and expert discussions on topics ranging from robotic surgery to fair data practices to generative AI's effect on creative work.
What Emerged From the Conversations
Three consistent themes surfaced across sessions. First, academia, industry, and government need closer collaboration on AI development. Second, critical thinking and continuous learning matter more as AI tools evolve. Third, researchers and policymakers must define what human responsibility means when robots and AI agents make decisions.
Peter Stone, chair of UT's Computer Science department, opened the symposium by noting the moment's significance. "Machine learning, artificial intelligence - especially generative artificial intelligence - is changing the world in many ways," Stone said. "This is the first time these three organizations have come together to give a joint symposium, but I think, especially in this moment, it's fitting to have machine learning, robotics and Good Systems Ethical AI all together in this room."
Research Computing and Open-Source Models
Adam Klivans, director of the Machine Learning Lab, emphasized the value of open-source AI models for responsible research. UT has access to the largest compute cluster in academia at the Texas Advanced Computing Center, which allows researchers to train models comparable to those developed by closed-source companies.
"We'll be able to train close to frontier-size models and try out a lot of ideas to find out what's really going on at these closed-source companies," Klivans said. The lab is developing open-source models through collaborations with the Institute for Foundations of Machine Learning and the Center for Generative AI.
Robotics in Healthcare and Hospital Design
José del R. Millán, director of Texas Robotics, highlighted UT's emerging academic health system as a testing ground for AI and robotics integration. The university is building a state-of-the-art, digitally enabled hospital where researchers can implement and study how intelligent systems function in clinical settings.
"We have the possibility to start thinking about and start implementing how artificial intelligence, how robotics will be integrated into that future hospital," Millán said.
Ethics and Human-Centered Approaches
Ken Fleischmann, director of Good Systems, stressed the need for values-driven AI adoption in government and national security contexts. "There's never been a more important time to question - what should we do?" Fleischmann said.
The symposium featured keynotes on fair data practices, physical AI capabilities, and language model advances. Alice Xiang, global head of AI Governance at Sony Group, presented Sony AI's Fair Human-centric Image Benchmark, which obtained consent from data contributors, compensated them, and allowed opt-outs from image training sets.
Gregory Hager from Johns Hopkins University discussed General Physical Intelligence - the ability for robots to understand and execute novel physical tasks - and how academia, industry, and government collaboration can advance the field. Kilian Weinberger from Cornell University presented a new class of language models that use parameters as computation and prioritize external memory.
What's Next
Stone expressed cautious optimism about technology's direction. "We have the opportunity to choose what technology we build and also try to shape it in a way that the positives will outweigh the negatives, and I'm optimistic that that's possible," he said.
UT's three research groups plan to continue advancing AI through courses, teaching practices, and training for the next generation of leaders. Generative AI and LLM Courses and AI Research Courses can help professionals stay current with these developments.
Recorded sessions from the symposium are available on YouTube. More information is available at ai.utexas.edu.
Your membership also unlocks:
