Vision Statement
- Conduct rigorous scientific research towards understanding various facets of biological intelligence
- Guide the development of artificial-intelligence hardware and software in a safe and ethical manner
- Openly disseminate the knowledge to academia, industries, and the general-public
Strategic Objectives
- Understanding biological neural systems to inspire future AI hardware and software.
- Developing robust theoretical frameworks and experimentally testable hypotheses.
- Creating rigorous benchmarks highlighting unique human and animal capabilities to drive AI advancement.
- Disseminating knowledge broadly to foster ethical and responsible research and technology development.
Core Themes
Neural Hardware and AI Architectures
Translating biological neural circuitry insights into novel hardware designs for AI systems.
Translating biological neural circuitry insights into novel hardware designs for AI systems.
Sensory Information Integration
Investigating how biological sensory systems integrate complex information, informing better sensory processing models in AI.
Investigating how biological sensory systems integrate complex information, informing better sensory processing models in AI.
Mesoscale Neural Dynamics
Examining rhythmic activities, sub-threshold dynamics, and spiking behaviors in biological brains to inform computational paradigms.
Examining rhythmic activities, sub-threshold dynamics, and spiking behaviors in biological brains to inform computational paradigms.
Developmental Trajectories of Intelligence
Studying intelligence progression from early development through adulthood in biological organisms, guiding adaptive and scalable AI designs.
Studying intelligence progression from early development through adulthood in biological organisms, guiding adaptive and scalable AI designs.
Human and Animal Behavioral Benchmarking
Identifying tasks uniquely mastered by biological organisms—such as causal reasoning, physical intuition, and context-awareness—to establish comprehensive benchmarks for AI evaluation.
Identifying tasks uniquely mastered by biological organisms—such as causal reasoning, physical intuition, and context-awareness—to establish comprehensive benchmarks for AI evaluation.
Research Methodology
- Theoretical Foundations: Rigorous theoretical modeling informed by existing neuroscientific data.
- Hypothesis-Driven Experiments: Designing experiments that yield tangible, verifiable outcomes in animal and human studies.
- Benchmarking and Validation: Establishing extensive, task-specific benchmarks emphasizing biological intelligence strengths for systematic AI evaluation.
- Innovative AI Architecture Development: Applying theoretical insights and experimental validations directly to the design of advanced AI systems.
Guiding Principles
- Focused Inquiry: Projects must clearly define and investigate one central question with precise hypotheses and identified null alternatives.
- Testability and Practicality: All theoretical frameworks must produce tangible, experimentally testable predictions, answering the critical question: "So what?"
- Alignment with Existing Knowledge: Clearly articulate how new theories complement, extend, or challenge existing well-established models and results.
- Impact on Consciousness Research: Directly address the conceptual implications of perception and consciousness, clarifying theoretical debates and informing practical approaches in clinical and cognitive research.
- Ethical and Societal Responsibility: Emphasize responsible and ethically grounded applications, ensuring alignment with broader societal goals.
Impact and Invitation
Saphe Labs seeks measurable impacts by transforming neuroscience understanding and guiding ethical AI development. We invite researchers, innovators, and collaborators to join us in reshaping the future of intelligence research and technology.