AIAS⁺ 2026 concluded on November 7 with fresh perspectives and important questions still to explore. The final day’s discussions spanned human health, the future of research and education, and the challenge of making scientific findings trustworthy.
@DaphneKoller opened the day with her morning plenary, Causal AI for Human Health.
The AI & Mental Health track featured 2026 Nobel laureate @KarlDeisseroth’s lecture on optical tools for studying the brain. Shannon Wiltsey Stirman explored AI tools for therapist training and between-session support, while @KafuiDzirasa presented work on mapping and editing neural circuits for social behavior. Christina Kim @tinakim_neuro discussed AI-guided protein engineering to map psychedelic action in the brain.
@USC President @BeongKim brought the discussion into higher education with University 3.0: Reimagining Research and Education in the Age of AI. He explored personalized AI tutoring and new research partnerships, while emphasizing the importance of independent thinking and human development.
In Discoverative AI, @SimonShaoleiDu’s Deep Discovery for Future Prediction explored forecasting as a testing ground for AI reasoning, causal discovery and self-correction. KyungHyun Cho @kchonyc examined the connection between computational design and experimental validation.
The afternoon’s Human–AI Co-discovery sessions featured @WillettBecca on uncertainty quantification, alongside @james_y_zou, @lecong and @SGRodriques on AI agents, programmable medicines and scientific discovery.
Michael Jordan examined bias in scientific AI, while James J. Collins discussed deep learning for antibiotic discovery. The final panel brought verification, reliability and public trust into focus.
We also celebrated the next generation of researchers at the Student Poster Session. Congratulations to our award recipients:
🏆 AI for Science Poster: Zixuan (Helen) Liu
🏆 Best Poster: David Li and Isabella Li
Thank you to every student who shared their work and contributed to the symposium.
In her closing remarks, Chrissy Luo emphasized that finding an answer is not the same as understanding it, and identifying a promising pattern is not the same as knowing its cause. The AI for Science Pulse Poll reflected this tension: 71% of respondents pointed to verification as the likely bottleneck if AI made research ten times faster. As discovery accelerates, strengthening the tools, methods and standards that make findings trustworthy remains essential.
Looking ahead, Chrissy announced that Nobel laureates Carolyn Bertozzi and @francesarnold will join AIAS⁺ 2027 as plenary lecturers.
Thank you to our speakers, students, program committee, partners and attendees for three days of thoughtful discussion and collaboration. See you next year!
#AIASPlus26 #AIForScience #ChenInstitute
@DaphneKoller opened the day with her morning plenary, Causal AI for Human Health.
The AI & Mental Health track featured 2026 Nobel laureate @KarlDeisseroth’s lecture on optical tools for studying the brain. Shannon Wiltsey Stirman explored AI tools for therapist training and between-session support, while @KafuiDzirasa presented work on mapping and editing neural circuits for social behavior. Christina Kim @tinakim_neuro discussed AI-guided protein engineering to map psychedelic action in the brain.
@USC President @BeongKim brought the discussion into higher education with University 3.0: Reimagining Research and Education in the Age of AI. He explored personalized AI tutoring and new research partnerships, while emphasizing the importance of independent thinking and human development.
In Discoverative AI, @SimonShaoleiDu’s Deep Discovery for Future Prediction explored forecasting as a testing ground for AI reasoning, causal discovery and self-correction. KyungHyun Cho @kchonyc examined the connection between computational design and experimental validation.
The afternoon’s Human–AI Co-discovery sessions featured @WillettBecca on uncertainty quantification, alongside @james_y_zou, @lecong and @SGRodriques on AI agents, programmable medicines and scientific discovery.
Michael Jordan examined bias in scientific AI, while James J. Collins discussed deep learning for antibiotic discovery. The final panel brought verification, reliability and public trust into focus.
We also celebrated the next generation of researchers at the Student Poster Session. Congratulations to our award recipients:
🏆 AI for Science Poster: Zixuan (Helen) Liu
🏆 Best Poster: David Li and Isabella Li
Thank you to every student who shared their work and contributed to the symposium.
In her closing remarks, Chrissy Luo emphasized that finding an answer is not the same as understanding it, and identifying a promising pattern is not the same as knowing its cause. The AI for Science Pulse Poll reflected this tension: 71% of respondents pointed to verification as the likely bottleneck if AI made research ten times faster. As discovery accelerates, strengthening the tools, methods and standards that make findings trustworthy remains essential.
Looking ahead, Chrissy announced that Nobel laureates Carolyn Bertozzi and @francesarnold will join AIAS⁺ 2027 as plenary lecturers.
Thank you to our speakers, students, program committee, partners and attendees for three days of thoughtful discussion and collaboration. See you next year!
#AIASPlus26 #AIForScience #ChenInstitute
