AI and neuroscience have long evolved in tandem — early AI drew inspiration from the brain, and modern AI has now reached a capability that can fundamentally reshape how neuroscience is done.
This workshop explores a phase in which AI moves beyond data analysis and modeling to become an active participant in scientific discovery. We take inspiration from the history of neuroscience — how transformative ideas about the brain were conceived, tested, and refined — and build toward a more ambitious question:
Can AI systems learn to generate and pursue scientific insight at superhuman scale?
We envision discoverative AI systems that do not merely process data, but continuously formulate hypotheses, design experiments, generate data, and iteratively refine their understanding of neural systems — shifting from stochastic, human-expert-limited hypothesis generation to expansive, automated exploration of the brain's complexity.
How this differs from brain-inspired AI
Where NeuroAI efforts (e.g. at CSHL and NIH) build next-generation AI inspired by the brain, this workshop emphasizes the reverse direction: leveraging modern AI to advance neuroscience itself — systems that generate hypotheses, design experiments, and participate in iterative scientific discovery.
