A Paradigm for AI Consciousness
Seeds of Science June 1, 2024 DOI: 10.53975/i2e7-1t5x (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Theoretical or philosophical paper Peer reviewed |
|---|---|
| Key points | Argues that a physicalist, decoherence-based framework with symmetry considerations can ground knowledge about AI consciousness, and that brains and computers differ significantly in branchial space, with computational efficiency potentially mitigating 's-risk'. |
Abstract
How can we create a container for knowledge about AI consciousness? This work introduces a new framework based on physicalism, decoherence, and symmetry. Major arguments include (1) atoms are a more sturdy ontology for grounding consciousness than bits, (2) Wolfram’s ‘branchial space’ is where an object’s true shape lives, (3) electromagnetism is a good proxy for branchial shape, (4) brains and computers have significantly different shapes in branchial space, (5) symmetry considerations will strongly inform a future science of consciousness, and (6) computational efficiency considerations may broadly hedge against “s-risk”.