CTM-AI: A Blueprint for General AI Inspired by Consciousness

Haofei Yu, Yining Zhao, Lenore Blum, Manuel Blum, Paul Pu LiangSubmission to ICLR 2026 · Rejected

TL;DR

This paper operationalizes the Conscious Turing Machine (CTM)—a formal cognitive architecture inspired by Alan Turing’s computation model and Bernard Baars’ Global Workspace Theory—into an actionable AI system called CTM-AI. By replacing rigid, centralized agent workflows with competitive workspace dynamics, global broadcasting, and emergent inter-processor links, CTM-AI coordinates specialized multimodal models and tools without requiring a central orchestrator.

What Problem Does This Paper Address?

Current AI systems predominantly rely on two paradigms:

  1. Monolithic foundation models, which scale parameters centrally but lack modularity and struggle to incorporate new capabilities dynamically.
  2. Multi-agent orchestrations, which typically depend on rigid, pre-programmed pipelines or a centralized "conductor" agent to manage message passing.

In cognitive science, the Conscious Turing Machine (Blum & Blum) formalizes human-like consciousness without a central executive: specialized "unconscious" processors run in parallel, compete to place a summary of their state onto a limited-capacity "conscious" stage (short-term memory), and broadcast that winner back to the collective. Until now, CTM remained purely theoretical. This paper translates CTM’s formal mathematical abstraction into a practical, modular software framework powered by modern foundation models.

Methods

CTM-AI implements CTM’s formal 7-tuple architecture using standard LLM and multimodal backbones (primarily gemini-2.0-flash-lite):

[Parallel LTM Processors] ──(Up-Tree Competition)──> [STM Workspace (Winner)]
           ▲                                                        │
           └────────────(Down-Tree Broadcast)───────────────────────┘
                                   │
              (Dynamic Link Formation & Multimodal Fusion)
  • Long-Term Memory (LTM) Processors: Specialized modules (sensory decoders, external APIs/tools, cognitive reasoners) that run in parallel. At step $t$, each generates a chunk: a tuple containing a natural-language gist, an information query for other processors, and a self-assessed utility weight (derived from relevance, confidence, and surprise).
  • Up-Tree Competition & Short-Term Memory (STM): The STM holds only a single chunk. Instead of an explicit orchestrator assigning tasks, an up-tree competition selects the winning chunk with the highest confidence/relevance weight. A supervisor LLM checks whether the query is resolved.
  • Down-Tree Broadcast: The winning chunk is broadcast globally to all LTM processors, causing the entire collective to attend to the winner’s findings and update their private memory.
  • Unconscious Link Formation: If two processors repeatedly find each other’s responses relevant (relevance score $>0.8$), a bidirectional link forms; if relevance drops ($<0.2$), the link decays.
  • Multimodal Fusion: Linked processors directly query each other in parallel ("unconscious communication") to resolve mutual questions before entering the next competitive cycle.

Datasets

The authors evaluated CTM-AI zero-shot across multiple existing benchmarks (no new datasets were introduced):

  • Multimodal Perception & Social Semantics: MUStARD (multimodal sarcasm detection), UR-Funny (multimodal humor detection), and NYCartoon (New Yorker caption contest).
  • Tool Learning & API Calling: StableToolBench (multi-turn tool invocation via the MirrorAPI server).
  • Web Agents: A 40-task subset of WebArena (OneStopShop domain).

Key Insights and Ideas

  • Decentralized Coordination: The system eliminates the "central executive" bottleneck. Information routing emerges organically from competitive bidding and global broadcasting.
  • Unconscious Efficiency: Once two processors establish a link, they share context laterally without cluttering the global workspace, mirroring human automatization (e.g., coordinating sight and motor skills when riding a bicycle).
  • Generalizing Reasoning: Single-processor CTM-AI reduces naturally to standard chain-of-thought ("o1-style") reasoning, while multi-processor setups generalize multimodal multi-agent workflows within a single formalism.

Key Results

  • Sarcasm Detection (MUStARD): CTM-AI achieved 73.88% Accuracy and 73.77 F1, outperforming the Gemini base model (70.42% Acc / 70.26 F1) and specialized models like MMoE (70.41% Acc).
  • Tool Learning (StableToolBench): CTM-AI marked substantial improvements in Solvable Pass Rate over both the Base Model + CoT and GPT-4o baselines:
  • I2-Category: 39.1% (vs. 26.3% Base, 32.8% GPT-4o DFS)
  • I2-Instruction: 51.5% (vs. 37.2% Base, 28.3% GPT-4o DFS)
  • I3-Instruction: 38.5% (vs. 18.5% Base, 23.0% GPT-4o DFS)
  • Web Navigation (WebArena): Succeeded on 8/40 tasks compared to 6/40 for standard LLM baselines.
  • Ablation Studies: Removing Long-Term Memory (F1 dropped from 73.77 to 64.59) or down-tree broadcasting (F1 dropped to 65.12) caused the steepest degradation, showing that persistent memory and global attention are critical to CTM-AI's reasoning.

Limitations and Future Work

  • Parallelism Scale: While theoretical CTM supports millions of processors, the practical implementation tests only a small handful ($<10$ processors), replacing hierarchical binary-tree tournament selection with a simplified global argmax.
  • Misleading Signals & Cascading Inquiries: On NYCartoon (54.7% matching vs. 59.7% base) and failed cases in UR-Funny, partial visual inputs misled the model. The LTM processors looped on unanswerable questions (e.g., repeatedly demanding missing facial expressions), creating spurious links that degraded performance.
  • Tool Hallucination: Processors occasionally judged that they could not invoke an API without actually issuing the tool call.
  • Future Directions: Expanding processor counts to require full hierarchical up-trees, incorporating learned processor representations, and mitigating confirmation bias across modalities.

Summary updated

Abstract

Despite remarkable advances, today's AI systems remain narrow in scope, falling short of the flexible, adaptive, and multisensory intelligence that characterizes humans. This gap has fueled longstanding debates about whether AI might one day achieve human-like generality or even consciousness, and whether principles of consciousness can inspire new architectures for general AI. This paper presents an early blueprint for implementing a general AI system based on the Conscious Turing Machine (CTM), a formal machine model of consciousness. The CTM has an enormous number of powerful processors ranging from specialized experts (e.g., vision–language models, search engines, APIs) to unspecialized general-purpose learners poised to develop their own expertise. Crucially, for whatever problem must be dealt with, the system need not know in advance which processors hold the relevant expertise; instead, multimodal machine learning methods enable the system to select, integrate, and fuse information across processors. We extend the CTM into a practical framework, the CTM-AI, and demonstrate its utility on diverse tasks including multimodal perception, tool learning with multiple APIs, and multi-turn web agent tasks. Together, this work offers a principled and testable blueprint for general AI inspired by computational models of consciousness.
BibTeX
@misc{yu2026ctm,
  author = {Haofei Yu and Yining Zhao and Lenore Blum and Manuel Blum and Paul Pu Liang},
  title = {{CTM-AI: A Blueprint for General AI Inspired by Consciousness}},
  year = {2026},
  eprint = {2605.04097},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.04097},
}
Figure