indicator properties of consciousness
In plain wordsA checklist of design features that leading scientific theories link to consciousness, used to judge how likely it is that a particular AI system could be conscious.
Properties of a system's architecture or processing that scientific theories of consciousness associate with conscious experience, used to assess whether an AI system is a candidate for machine consciousnessmachine consciousness, n.: Whether a machine such as an AI system could have genuine experiences of its own — and the research field that tries to find out, or to build such a machine. without relying on its own testimony. The influential 2023 report by Patrick Butlin, Robert Long and 17 co-authors, which adopts computational functionalismfunctionalism, n.: The view that minds are defined by how they work, not what they are made of: pain, for example, is whatever inner state plays pain's part, linking injury, other thoughts and behavior. as a working hypothesis, lists fourteen indicators: twelve derived from global workspace theoryglobal workspace theory, n.: A leading scientific theory holding that we become conscious of something when the brain broadcasts it widely — as if putting it on a stage where every part of the mind can use it., recurrent processing theoryrecurrent processing theory, n.: A theory that we consciously see something only when signals in the brain's visual areas loop back on themselves, not when they merely sweep forward once., higher-order theorieshigher-order theories of consciousness, n.: Theories holding that a mental state is conscious when the mind, in effect, notices it — when some further thought or perception represents you as being in that state., predictive processingpredictive processing, n.: The idea that the brain is a prediction machine: it constantly guesses what its senses will report next and learns mainly from its mistakes, so what we perceive is largely its best guess. and attention schema theoryattention schema theory, n.: A theory that the brain keeps a simplified sketch of its own attention, and that our feeling of having an inner awareness is simply how that sketch portrays attention., and two concerning agency and embodiment. Examples include a limited-capacity workspace with global broadcast, and metacognitive monitoring that can tell reliable perceptual signals from noise. The more indicators a system has, the more credence its consciousness deserves. The authors judged that no AI system then existing was conscious, but saw no evident technical obstacle to building systems that have the indicators.
Origin
Set out, with the 'theory-heavy' method it implements, in 'Consciousness in Artificial Intelligence: Insights from the Science of Consciousness' (arXiv, August 2023) by Patrick Butlin, Robert Long and colleagues including Yoshua Bengio, Jonathan Birch and Eric Schwitzgebel. Research on animal sentience had earlier used behavioral and neural markers in a similar way.
Usage note
The method is only as reliable as the theories it draws on, none of which is established, and it presupposes computational functionalism, which biological naturalismbiological naturalism, n.: John Searle's view that consciousness is a real biological product of brains, as digestion is of stomachs, so running a computer program would not by itself create a mind. and integrated information theoryintegrated information theory, n.: A scientific theory that measures consciousness by how much a system works as a single, unified whole rather than as separate parts. It implies that ordinary computers, whatever software they run, have little or no consciousness. reject. By design it sidesteps behavioral tests, which language modelslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. can pass by imitation.
Related
- machine consciousnessmachine consciousness, n.: Whether a machine such as an AI system could have genuine experiences of its own — and the research field that tries to find out, or to build such a machine.
- global workspace theoryglobal workspace theory, n.: A leading scientific theory holding that we become conscious of something when the brain broadcasts it widely — as if putting it on a stage where every part of the mind can use it.
- higher-order theories of consciousnesshigher-order theories of consciousness, n.: Theories holding that a mental state is conscious when the mind, in effect, notices it — when some further thought or perception represents you as being in that state.
- neural correlates of consciousnessneural correlates of consciousness, n.: The particular brain activity that reliably goes with a conscious experience — the minimum the brain must be doing for you to see, hear or feel something.
- AI welfareAI welfare, n.: Concern for the well-being of AI systems themselves, in case some of them can suffer or have interests that matter — and the research into how to tell and what to do about it.
- moral patienthoodmoral patienthood, n.: Being the kind of being whose interests matter morally for its own sake, so that it can be wronged. Babies and animals count, even though they are not responsible for what they do.
Sources
- Butlin, P., Long, R., Elmoznino, E., Bengio, Y., Birch, J. et al. (2023). Consciousness in Artificial Intelligence: Insights from the Science of Consciousness. arXiv:2308.08708. arxiv.org/abs/2308.08708
Mentioned in
AI welfareAI welfare, n.: Concern for the well-being of AI systems themselves, in case some of them can suffer or have interests that matter — and the research into how to tell and what to do about it., consciousnessconsciousness, n.: Having experiences: there being something it feels like to be you, as when you see red or feel pain. Whether any AI is or could become conscious is an open and fiercely debated question., David ChalmersDavid Chalmers, prop. n.: An Australian philosopher who named the 'hard problem' of consciousness — why physical processes feel like anything at all — and who asks whether AI systems could be conscious., functionalismfunctionalism, n.: The view that minds are defined by how they work, not what they are made of: pain, for example, is whatever inner state plays pain's part, linking injury, other thoughts and behavior., global workspace theoryglobal workspace theory, n.: A leading scientific theory holding that we become conscious of something when the brain broadcasts it widely — as if putting it on a stage where every part of the mind can use it., machine consciousnessmachine consciousness, n.: Whether a machine such as an AI system could have genuine experiences of its own — and the research field that tries to find out, or to build such a machine., attention schema theoryattention schema theory, n.: A theory that the brain keeps a simplified sketch of its own attention, and that our feeling of having an inner awareness is simply how that sketch portrays attention., higher-order theories of consciousnesshigher-order theories of consciousness, n.: Theories holding that a mental state is conscious when the mind, in effect, notices it — when some further thought or perception represents you as being in that state., neural correlates of consciousnessneural correlates of consciousness, n.: The particular brain activity that reliably goes with a conscious experience — the minimum the brain must be doing for you to see, hear or feel something., recurrent processing theoryrecurrent processing theory, n.: A theory that we consciously see something only when signals in the brain's visual areas loop back on themselves, not when they merely sweep forward once.