community
active
leiden_hybrid_concepts
label: haiku
community:leiden_hybrid_concepts-run4-c6-c5Global workspace theory as AI consciousness framework
Evaluating whether current architectures (transformers, perceivers) satisfy GWT's indicator properties for machine consciousness.
6 members. Each node is clickable.
Loading graph…
Drawn from 2 sources
The papers/notes whose extracted claims & findings make up this cluster.
- Consciousness in Artificial Intelligence: Insights from the Science of Consciousness5 members
- 2026-05-14_phil-trans-A-goodfire-aboutblank-impact.md1 member
Bridges (1)
Other communities that share members with this one — cross-cutting threads or papers that sit at the seam between two themes.
Claims (6)
- Many of the indicator properties can be implemented in AI systems using current techniques.Feasibility demonstrated in Section 3.1.
- No current AI systems are conscious.Overall assessment of current state; qualified by possibility.
- Perceiver architecture is some way from being a full implementation of GWT.Case study conclusion.
- There are no obvious technical barriers to building AI systems which satisfy the indicator properties.Feasibility claim about near-term conscious AI.
- Transformer-based LLMs do not clearly satisfy the GWT indicator properties.Assessment from case study.
- Current AI systems suffer from temporal locked-in syndrome lacking time-awareness design.