# Consciousness and AI

Whether an AI can have subjective experience is a central philosophical question. This sub-article expands the consciousness discussion in [Philosophy](../philosophy.md).

## Philosophical Positions
- **Functionalism**: consciousness is about causal role, so AI could be conscious.
- **Biological naturalism** (Searle): consciousness requires biological causation.
- **Panpsychism** and **information integration** theories (IIT) give different criteria.

## The Hard Problem
- Explaining why physical states produce subjective experience ("what it is like").
- If a model perfectly imitates a conscious human, is it conscious?

## Practical Relevance
- If AI could be conscious, questions of rights, welfare, and suffering arise.
- Most current systems are far from consensus on these criteria.

## Open Questions
- Do LLMs have beliefs? Do they have qualia? Do they "want" anything?
- No agreed test for machine consciousness exists.

## Sources
- Chalmers, D. (1996). *The Conscious Mind.*
- Tononi, G. (2012). *Integrated Information Theory of Consciousness.*

*Back to [Philosophy](../philosophy.md).*