Noumenal Labs White Paper: How To Build A Brain

TL;DR

This work addresses the grounding problem by arguing that robust AI should be grounded in world-structured, object-centered representations rather than word-based embeddings. Grounded world models, grounded in Bayesian mechanics and the free energy principle, are proposed to be compositional, predictive yet explanatory, and capable of autonomous causal discovery through active interventions. The authors outline a framework built on object boundaries via Markov blankets, sparse relational dynamics, and a common language for interactions, enabling scalable reasoning and generalization. They further discuss the design of agents that behave like scientists and the practical benefits for 3D world modeling, simulation, and domain-specific causal analysis, aiming for AI that augments human understanding and supports systems engineering progress.

Abstract

This white paper describes some of the design principles for artificial or machine intelligence that guide efforts at Noumenal Labs. These principles are drawn from both nature and from the means by which we come to represent and understand it. The end goal of research and development in this field should be to design machine intelligences that augment our understanding of the world and enhance our ability to act in it, without replacing us. In the first two sections, we examine the core motivation for our approach: resolving the grounding problem. We argue that the solution to the grounding problem rests in the design of models grounded in the world that we inhabit, not mere word models. A machine super intelligence that is capable of significantly enhancing our understanding of the human world must represent the world as we do and be capable of generating new knowledge, building on what we already know. In other words, it must be properly grounded and explicitly designed for rational, empirical inquiry, modeled after the scientific method. A primary implication of this design principle is that agents must be capable of engaging autonomously in causal physics discovery. We discuss the pragmatic implications of this approach, and in particular, the use cases in realistic 3D world modeling and multimodal, multidimensional time series analysis.