NethoLabs

AI-Neuroethology, Whole-Brain Emulation and AI Safety

AI-driven neuroethology is offering unprecedented opportunities to advance neuroscience: collection and analysis of data spanning weeks to years — at the PetaByte (PB) scale - while helping to significantly decrease the number of research animals required. We can vastly improve the power of our studies while gaining new insights into anatomical and functional circuits involved in brain disorders.

Whole-Brain Emulation (WBE) as a breakthrough scientific tool. AI-driven paradigms support the science of whole-brain dynamical models such as whole-brain emulators (WBE): i.e., measuring, modeling, and emulating entire nervous systems of mammals and testing them in digital environments. WBE paradigms are theoretically and experimentally sound and provide a new lenses into neural dynamics from the whole-organism perspective.

WBEs advance AI safety. WBEs emulate biological brains in interpretable, ground-truth conditions - while still running on known architectures such as transformer-based neural networks (NNs). Thus WBEs offer a radically transparent substrate for AI alignment with better-understood anatomy and behaviour, enabling causal insights into goal formation, agency, and failure modes. This makes WBEs important tools for determining when advanced AI systems may become critically unsafe for humanity.