AI Is Accelerating Scientific Discovery. What Will It Take to Trust the Results?
REDWOOD CITY, Calif., Nov. 18, 2026
Press Release Disclaimer: This is a press release distributed through the XPR Media network. It has not been independently verified by our newsroom.

![]()
AI Is Accelerating Scientific Discovery. What Will It Take to Trust the Results?
PR Newswire
REDWOOD CITY, Calif., Nov. 18, 2026
At AIAS+ 2026, leading researchers explored advances in drug discovery, mathematics and neuroscience-and the evidence and human judgment needed to turn AI’s promise into meaningful progress.
REDWOOD CITY, Calif., Nov. 18, 2026 /PRNewswire/ — Artificial intelligence is helping scientists identify potential antibiotics, design new materials and explore complex mathematical problems. But as AI accelerates research, a fundamental challenge remains: How do scientists determine which AI-generated results are reliable, reproducible and ultimately useful?
That question emerged as a central theme at AIAS+ 2026: AI Advancing Science & Society, convened November 5–7 in San Francisco by The Tianqiao and Chrissy Chen Institute. In a voluntary poll of conference participants, 71% of respondents identified verifying results as the next bottleneck if AI made research ten times faster. Another 73% selected independent scientists who reproduce a discovery as the final authority on its credibility. The findings reflect participating respondents’ views, not a representative scientific survey.
The challenge is already apparent in drug discovery. MIT biological engineering professor James J. Collins described how AI can search vast molecular libraries for potential antibiotics, while laboratory testing remains a critical constraint.
“The challenge is that AI is outpacing our ability to create molecules that we can test in the lab,” Collins told the Chen Institute.
Watch the AIAS+ 2026 conference recap: https://youtu.be/vKsUskEUA6k?is=eHWz0LEk79p9NY-k
Expanding scientific possibilities while strengthening the evidence
Across the three-day symposium, researchers examined how AI is changing scientific discovery and the role of human expertise.
Daphne Koller, founder and CEO of insitro, discussed combining clinical data with cell experiments to identify potential drug targets. Washington University in St. Louis chemist Zhiling “Zach” Zheng explored AI-assisted materials discovery, while Stanford neuroscientist and Nobel Laureate, Karl Deisseroth examined experimental approaches to distinguishing correlation from causation.
In mathematics, UCLA Fields Medalist Terence Tao described how AI could help researchers investigate more problems and identify promising directions, while emphasizing the distinction between generating answers and developing genuine understanding.
“AIs have become very good at producing answers, but not great at producing understanding,” Tao told The Chen Institute.
Chen Institute co-founder Tianqiao Chen emphasized in his special lecture that verification is essential to determining whether AI-generated hypotheses stand up as credible scientific discoveries. He called for infrastructure that combines independent evidence, experimental testing and human scientific judgment to keep pace with AI-driven research.
Together, the discussions underscored that different scientific disciplines require different methods for establishing confidence in results, from mathematical proofs to laboratory experiments and clinical evaluation.
Putting human needs at the center of scientific progress
Speakers also examined how AI-assisted research can translate into meaningful benefits for people.
UC Davis neuroscientist Sergey Stavisky, recipient of the 2026 Chen Institute and Science Prize for AI Accelerated Research, discussed AI-assisted speech restoration aimed at preserving a person’s intended words and expression. Researchers from Dartmouth and Stanford also explored applications in mental health care, emphasizing clinical evaluation, patient safety and treatment outcomes.
“That’s why we need to strengthen the tools, methods, and standards that make scientific discoveries trustworthy,” said Chrissy Luo, co-founder of the Tianqiao and Chrissy Chen Institute, in her closing remarks.
Looking ahead to AIAS+ 2027
AIAS+ 2026 received 102 research-paper submissions and accepted 58 following peer review involving 127 reviewers. The conference also featured research presentations, mentorship opportunities and a student showcase connecting emerging researchers with established scientists.
AIAS+ 2027 will take place October 21–23, 2027, in San Francisco, with Nobel laureates Carolyn Bertozzi of Stanford University and Frances Arnold of Caltech confirmed as plenary lecturers.
About the Tianqiao and Chrissy Chen Institute
The Tianqiao and Chrissy Chen Institute supports scientific research and collaboration aimed at advancing understanding of the brain, accelerating discovery and improving human well-being. Through research programs, scientific partnerships and interdisciplinary convenings such as AIAS+, the Institute brings together scientists and innovators to address important questions at the intersection of science, technology and society.
View original content to download multimedia:https://www.prnewswire.com/news-releases/ai-is-accelerating-scientific-discovery-what-will-it-take-to-trust-the-results-302943295.html
SOURCE Tianqiao & Chrissy Chen Institute



