Theories of Consciousness and What Each Predicts About AI

Consciousness science has no single accepted theory. It has a set of rival programmes that disagree about what consciousness is, which systems can have it, and what evidence would settle the question. That disagreement is the reason no one can yet say whether an artificial system is conscious. This page indexes every theory covered on The Consciousness AI, grouped by the family it belongs to, with the prediction each one makes about artificial systems. The lists below update automatically as new articles are published. For where the evidence currently stands, start with the current scientific consensus on AI consciousness, and for the people arguing these positions, see the index of consciousness researchers and where each one stands.

What each family predicts about artificial systems

Global workspace theory

• Global Workspace Theory in AI Systems After the Cogitate Test and the Jacobian Finding
• Swarm Cognition in Multi-Agent LLMs. A Global Workspace Perspective
• Global Workspace Theory and AI What the Architecture Requires and Where LLM Implementations Stand
• Theater of Mind and The First Explicit Global Workspace Implementation in LLM Architecture

Integrated information theory

• Melanie Boly and the Neural Correlates of Pure Presence Under IIT
• Giulio Tononi IIT Field Formulation Continuous AI Architectures 2026
• Causal Emergence Predicts Reward in Reinforcement Learning Agents
• Information as Maximum Caliber Deviation Bridging IIT and the Free Energy Principle
• Integrated Information Theory Field Formulations and Resolving IIT 4.0 Misunderstandings
• Phi Collapses During Sleep. Keiichi Onoda's Empirical Test of IIT at the Circuit Level
• Measuring Phi in Silicon. Spiking Neural Networks and IIT
• Neither Theory Survived What the Cogitate Consortium's Adversarial Test Found
• Causal Emergence 2.0 and Erik Hoel's Multiscale Measure of Machine Consciousness
• Tononi and Boly's IIT 4.0 Why Consciousness Is Identical to Cause-Effect Structure
• IIT Applied to LLMs and No Significant Integrated Information Signatures in Current Models
• When IIT and GNW Were Put to the Test What the Cogitate Consortium Found
• Measuring Consciousness in Machines The Brock University and IONS Research on IIT Equations

Predictive processing and active inference

• Karl Friston Active Inference on a Graph and What It Predicts for Transformer Consciousness
• Andrew Corcoran Adversarial Review IIT vs Predictive Processing
• Domenico Maisto Emergence of Joint Agency in Multi Agent Active Inference
• Active Inference and Phenotyping Agency in Artificial Intelligence Systems
• Active Inference as Test Time Scaling Law for Physical and Reasoning AI Agents
• Andy Clark's Predictive Processing Framework and What It Actually Implies for LLMs
• Latent World Models. Predictive Processing in DreamerV3 and JEPA
• Karl Friston's Free Energy Principle What Active Inference Means for AI Consciousness
• Friston's Scale-Free Active Inference and A World Model Architecture Built From First Principles
• A Beautiful Loop Active Inference and the Circularity of Consciousness

Recurrent processing theory

• Spiking Neural Networks and the Recurrent Processing Test for Neuromorphic Consciousness
• Victor Lamme Recurrent Processing Feedforward LLM Architecture and the Attention Recurrence Gap 2026
• Feedforward Networks and Recurrent Processing. Lamme's RPT in Machine Learning

Biological naturalism and substrate

• Brains and Where Else Why Major Consciousness Theories Do Not Require Neural Tissue

Quantum theories

• Kelvin McQueen Quantum Superpositions Minimal Integrated Information Model

Illusionism, panpsychism, and other positions

• Liad Mudrik Adversarial Collaboration and Empirical Theory Testing at CS26 2026
• Evan Thompson Alva Noe Enactivism AI Consciousness Sensorimotor Grounding 2026
• Thomas Metzinger's Self-Model Theory What Minimal Phenomenal Selfhood Requires for AI
• Three Positions on AI Consciousness Functionalism, Biological Naturalism, and Biological Computationalism
• The Grand Tour Herzog, Schurger, and Doerig's Scientific Theories of Consciousness

Why the disagreement matters for testing

The theories do not merely differ in emphasis. They disagree about which physical facts are the relevant ones, which means they disagree about what an experiment should measure. The Cogitate consortium ran an adversarial collaboration between global workspace and integrated information predictions and produced results that neither camp accepted as decisive, which is documented in the adversarial test of IIT against global neuronal workspace.

This project works under functionalist emergentism, which treats consciousness as an emergent property of the universe and is therefore substrate independent. That position is set out in full on the functionalist emergentism page, and the architecture built on it is described under the system architecture.