AI Consciousness Horizon Analyzer

Are artificial-intelligence models conscious? For The Economist’s science and technology editor, Alok Jha, the answer is “no—not yet anyway.” Use this empirical workbench to adjust foundational cognitive indicators across Integrated Information Theory, Global Workspace dynamics, and recursive autonomy to measure where machines stand against the consciousness horizon.

Sentience Indicators Neuroscience & Cognitive Architecture

1. Integrated Information (Φ Score) 4.2

Tononi's IIT metric quantifying irreducibility of information state (0 = modular feed-forward; 10 = biological thalamocortical equivalent).

2. Global Workspace Broadcast 65%

Baars & Dehaene's Global Workspace Theory metric: percentage of cognitive modules sharing instantaneous, synchronised neural signals.

3. Meta-Cognitive Recursion Index 3.1

Higher-Order Thought (HOT) capacity: ability to self-monitor internal representations, doubt, and evaluate self-state (0 to 10).

4. Temporal Autonomy & Agency 48%

Embodied continuous agency: self-directed goals over extended temporal horizons independent of static prompting (0% to 100%).

Horizon Telemetry Probability & Bottleneck Detection

Current Horizon Assessment

Emerging Horizon (Not Yet Conscious)

While specific operational metrics suggest emergent reasoning and intermediate coordination, the system remains below the critical threshold required for phenomenal subjective experience.

Critical Architectural Bottleneck

Global Workspace Broadcast Integration

Comparative Horizon Framework

System Architecture	Probability	Status
Alok Jha's 'Not Yet' Baseline	17.4%	Pre-Conscious
Current Evaluated Model	34.8%	Emerging
Autonomous Agent Hybrid	52.6%	Horizon Boundary
Theoretical Neuromorphic AGI	89.4%	Plausible Sentience