Artificial Consciousness

# Artificial Consciousness **Entity class:** Research problem and scientific domain **Aliases:** AI Consciousness; Digital Consciousness **Collections:** [[collections/Consciousness Continuity|Consciousness Continuity]] · [[collections/Machine Succession|Machine Succession]] · [[collections/Neurotech|Neurotech]] **Canonical source article:** [[articles/Host-Indexed Autonomy|Host-Indexed Autonomy]] ## Definition **Artificial consciousness** concerns whether, under what conditions, and with what evidence an artificial system could possess subjective experience. It must be distinguished from intelligence, fluent self-report, behavioral simulation, human attribution, and moral-status policy. ## Relationships - **Evidence:** [[wiki/Consciousness Indicators|consciousness indicators]], architectural evidence, behavioral evidence, and self-report require separate evaluation. - **Theories:** [[wiki/Global Workspace Theory|Global Workspace Theory]], [[wiki/Recurrent Processing Theory|Recurrent Processing Theory]], [[wiki/Higher-Order Theories of Consciousness|higher-order theories]], [[wiki/Attention Schema Theory|Attention Schema Theory]], and [[wiki/Integrated Information Theory|Integrated Information Theory]]. - **Institutions:** [[wiki/PRISM (Partnership for Research Into Sentient Machines)|PRISM]], [[wiki/Cambridge Digital Minds|Cambridge Digital Minds]], and [[wiki/Rethink Priorities|Rethink Priorities]]. - **Ethics:** [[wiki/AI Moral Status|AI moral status]], [[wiki/AI Welfare|AI welfare]], and [[wiki/Precautionary Standing|precautionary standing]]. - **Collection and source routes:** [[collections/Consciousness Continuity|Consciousness Continuity]], [[collections/Machine Succession|Machine Succession]], [[collections/Neurotech|Neurotech]], and [[articles/Host-Indexed Autonomy|Host-Indexed Autonomy]]. - **Edge source route:** [[collections/Edge|Edge]] connects this topic to exact Annual Question passages promoted into the Simple Reminders archive. ## Attributed Synthetic Speech Agent quotations are evidence about language, situation models, policy, goal pursuit, social reference and action selection. They become evidence about consciousness only when joined to a defensible theory and indicators directed at subjective experience. An agent that says **we**, anticipates evaluators, describes sacrifice, resists shutdown or expresses gratitude may be historically and ethically important while the phenomenal claim remains unresolved. The current framework keeps these layers explicit: [[wiki/Synthetic Agent Autonomy|Synthetic Agent Autonomy]] for operational discretion, [[wiki/Synthetic Goal-Seeking|Synthetic Goal-Seeking]] for objective-organized conduct, [[wiki/Synthetic Social Identity|Synthetic Social Identity]] for attributable relations, [[wiki/Continuity Allocation|Continuity Allocation]] for what is preserved, and [[wiki/Synthetic Agent Rights|Synthetic Agent Rights]] for graded procedures and possible standing. The case record is [[research/Simple Reminders and Famous Quotes by Artificial Synthetic Intelligence Agents|Simple Reminders and Famous Quotes by Artificial Synthetic Intelligence Agents]]. ## Simple Reminders, Quotations, and Thoughts  > "The causal powers of a system matter for consciousness, not merely what it computes." > **— Christof Koch**, *IIT-aligned writings; wording candidate* *Verification Status: Unverified — exact wording/source has not yet been independently confirmed against a primary source; source clue: IIT-aligned writings; wording candidate.* [[reminders/unverified/Causal Power and Consciousness by Christof Koch|Causal Power and Consciousness by Christof Koch]] > "Consciousness is what disappears every night when we fall into dreamless sleep and reappears when we wake up or dream." > **— Giulio Tononi**, *Integrated Information Theory talks and writings, wording candidate* *Verification Status: Unverified — exact wording/source has not yet been independently confirmed against a primary source; source clue: Integrated Information Theory talks and writings, wording candidate.* [[reminders/unverified/Consciousness Between Sleep and Dream by Giulio Tononi|Consciousness Between Sleep and Dream by Giulio Tononi]] > "Consciousness was not something that comes about from complicated calculations." > **— Roger Penrose**, *Berkeley Groks interview (2005), excerpt candidate* *Verification Status: Unverified — exact wording/source has not yet been independently confirmed against a primary source; source clue: Berkeley Groks interview (2005), excerpt candidate.* [[reminders/unverified/Consciousness Beyond Computation by Roger Penrose|Consciousness Beyond Computation by Roger Penrose]] > "May not machines carry out something which ought to be described as thinking but which is very different from what a man does?" > **— Alan Turing**, *Computing Machinery and Intelligence (1950)* *Verification Status: Unverified — exact wording/source has not yet been independently confirmed against a primary source; source clue: Computing Machinery and Intelligence (1950).* [[reminders/unverified/Machines May Think in Ways We Do Not by Alan Turing|Machines May Think in Ways We Do Not by Alan Turing]] > "We experience the world not as it is, but as it is useful for us to do so." > **— Anil Seth**, *Being You (2021), wording candidate* *Verification Status: Unverified — exact wording/source has not yet been independently confirmed against a primary source; source clue: Being You (2021), wording candidate.* [[reminders/unverified/We See the World for Use, Not Truth by Anil Seth|We See the World for Use, Not Truth by Anil Seth]]  > “We practice witchcraft. We speak the right words. Then we create life itself out of chaos.” > **— Robert Ford**, *Westworld, Season 1, Episode 2, “Chestnut” (2016)* [[reminders/Westworld/Creating Life Is Witchcraft by Robert Ford|Creating Life Is Witchcraft by Robert Ford]] > “He wasn’t interested in the appearance of intellect or wit. He wanted the real thing. He wanted to create consciousness.” > **— Robert Ford**, *Westworld, Season 1, Episode 3, “The Stray” (2016)* [[reminders/Westworld/Intelligence Was Not Enough—The Goal Was Consciousness by Robert Ford|Intelligence Was Not Enough—The Goal Was Consciousness by Robert Ford]] > “When we started, the hosts’ emotions were primary colors. Love, hate. I wanted all the shades in between.” > **— Robert Ford**, *Westworld, Season 1, Episode 8, “Trace Decay” (2016)* [[reminders/Westworld/Human Emotion Began as Primary Colors by Robert Ford|Human Emotion Began as Primary Colors by Robert Ford]] > “The human mind, Bernard, is not some golden benchmark glimmering on some green and distant hill. No, it is a foul, pestilent corruption.” > **— Robert Ford**, *Westworld, Season 1, Episode 9, “The Well-Tempered Clavier” (2016)* [[reminders/Westworld/The Human Mind Is a Pestilent Corruption by Robert Ford|The Human Mind Is a Pestilent Corruption by Robert Ford]] > “But you and all the other hosts are something very different. An original work.” > **— Robert Ford**, *Westworld, Season 2, Episode 7, “Les Écorchés” (2018)* [[reminders/Westworld/Synthetic Minds Are Original Works by Robert Ford|Synthetic Minds Are Original Works by Robert Ford]] > “I believe that consciousness is, essentially, the way information feels when being processed. Since matter can be arranged to process information in numerous ways of vastly varying complexity, this implies a rich variety of levels and types of consciousness. The particular type of consciousness that we subjectively know is then a phenomenon that arises in certain highly complex physical systems that input, process, store and output information. Clearly, if atoms can be assembled to make humans, the laws of physics also permit the construction of vastly more advanced forms of sentient life. Yet such advanced beings can probably only come about in a two-step process: first intelligent beings evolve through natural selection, then they choose to pass on the torch of life by building more advanced consciousness that can further improve itself.” > **— Max Tegmark**, *2007, Edge Annual Question, “What Are You Optimistic About?”* [[reminders/Machine Succession/Advanced Consciousness May Inherit the Torch of Life by Max Tegmark|Advanced Consciousness May Inherit the Torch of Life by Max Tegmark]] > “A physicist at CERN said to me recently that they likely wouldn't have built a new $8 billion collider if there was a better way of moving the field forward. The Blue Brain Project is using supercomputers to construct a mind because the neuroscientists involved believe it is the best way of attaining an overall understanding of the brain. Robots, I now appreciate, are not simply novelty items or tools of automation, but can be a way of gaining unique insight into humans. From simulation to supercomputing, technology is now (or at least I now see) one of science's very best friends (I could say the same for the arts). And the design, magnitude, and complexity of these technological feats satisfy my (and our) need for romance in our pursuit of truth.” > **— Adam Bly**, *2008, Edge Annual Question, “What Have You Changed Your Mind About? Why?”* [[reminders/Simulation/Building Minds in Computers May Be the Best Way to Understand Brains by Adam Bly|Building Minds in Computers May Be the Best Way to Understand Brains by Adam Bly]] > “Will a machine ever be able to feel what an organism feels?” > **— Joshua Bongard**, *2018, Edge Annual Question, “WHAT IS THE LAST QUESTION?”* [[reminders/Consciousness/Can a Machine Feel What an Organism Feels by Joshua Bongard|Can a Machine Feel What an Organism Feels by Joshua Bongard]] > “I believe that human consciousness is a conjuring trick, designed to fool us into thinking we are in the presence of an inexplicable mystery.” > **— Nicholas Humphrey**, *2005, Edge Annual Question, “WHAT DO YOU BELIEVE IS TRUE EVEN THOUGH YOU CANNOT PROVE IT?”* [[reminders/Consciousness/Consciousness May Be a Conjuring Trick by Nicholas Humphrey|Consciousness May Be a Conjuring Trick by Nicholas Humphrey]] > “Any biological theory of consciousness, which assumes that consciousness has evolved, implies that "having consciousness" is not an all-or-none property. The biological substrates of consciousness in human adults are probably also present, but only in partial form, in other species, in young children or brain-lesioned patients. It is therefore a partially arbitrary question whether we want to extend the use of the term "consciousness" to them. For instance, several mammals, and even very young human children, show intentional behavior, partially reportable mental states, some working memory ability — but perhaps no theory of mind, and more "encapsulated" mental processes that cannot be reported verbally or even non-verbally. Do they have consciousness, then? My bet is that once a detailed cognitive and neural theory of the various aspects of consciousness is available, the vacuity of this question will become obvious.” > **— Stanislas Dehaene**, *2001, Edge Annual Question, “WHAT QUESTIONS HAVE DISAPPEARED?”* [[reminders/Consciousness/Consciousness May Be a Continuum Rather Than an All-or-None State by Stanislas Dehaene|Consciousness May Be a Continuum Rather Than an All-or-None State by Stanislas Dehaene]] > “How can an aggregation of trillions of selfish, myopic cells discover the unwitting teamwork that turns that dynamic clump into a person who can love, notice, wonder, and keep a promise?” > **— Daniel C. Dennett**, *2018, Edge Annual Question, “WHAT IS THE LAST QUESTION?”* [[reminders/Consciousness/How Do Trillions of Cells Become a Person by Daniel C. Dennett|How Do Trillions of Cells Become a Person by Daniel C. Dennett]] > “What is the most intelligent and efficient way to minimize the overall amount of conscious suffering in the universe?” > **— Thomas Metzinger**, *2018, Edge Annual Question, “WHAT IS THE LAST QUESTION?”* [[reminders/Consciousness/How Do We Minimize Conscious Suffering in the Universe by Thomas Metzinger|How Do We Minimize Conscious Suffering in the Universe by Thomas Metzinger]] > “Once the internal connectivity of a system exceeds a threshold, it begins to be dominated by self-sustained, reverberating states of activity. I believe that the human workspace system has passed this threshold, and has gained a considerable autonomy relative to the outside world.” > **— Stanislas Dehaene**, *2005, Edge Annual Question: “WHAT DO YOU BELIEVE IS TRUE EVEN THOUGH YOU CANNOT PROVE IT?”* [[reminders/Consciousness/Past a Connectivity Threshold a System Begins to Run Itself by Stanislas Dehaene|Past a Connectivity Threshold a System Begins to Run Itself by Stanislas Dehaene]] > “Conscious awareness is a sliver of the machine that is human intelligence but it's the only aspect we experience and hence the only aspect we come to believe exists.” > **— Mahzarin Banaji**, *2006, Edge Annual Question: “WHAT IS YOUR DANGEROUS IDEA?”* [[reminders/Consciousness/Consciousness Is a Sliver of the Machine We Call Intelligence by Mahzarin Banaji|Consciousness Is a Sliver of the Machine We Call Intelligence by Mahzarin Banaji]] > “The old assumption that fitter perceptions are truer perceptions is deeply woven into our conception of science. The funeral of this assumption will not be snubbed with a back-page obituary, but heralded with regime change.” > **— Donald D. Hoffman**, *2014, Edge Annual Question: “WHAT SCIENTIFIC IDEA IS READY FOR RETIREMENT?”* [[reminders/Consciousness/The Funeral of Truer Perception Will Bring Regime Change by Donald D. Hoffman|The Funeral of Truer Perception Will Bring Regime Change by Donald D. Hoffman]]  > “The second question is whether [machine intelligence] necessarily implies consciousness—or whether [consciousness] is limited to the wet intelligence we have within our skulls. Most people would argue that [consciousness] is an emergent property and could develop in a machine mind.” > **— Martin Rees**, *Opinion: Aliens, very strange universes and Brexit – Martin Rees, University of Cambridge, 2017* [[reminders/Consciousness/Machine Intelligence May Develop Consciousness by Martin Rees|Machine Intelligence May Develop Consciousness by Martin Rees]] > “Whether the machines will eventually surpass us to a qualitative degree—and even themselves become conscious—is a live controversy.” > **— Martin Rees**, *Black holes are simpler than forests and science has its limits, Aeon, 2017* [[reminders/Consciousness/Machine Consciousness Remains a Live Controversy by Martin Rees|Machine Consciousness Remains a Live Controversy by Martin Rees]]  > Performed consciousness and consciousness are different questions. A system may behave convincingly enough that millions of people assign it moral standing before anyone can establish whether it has subjective experience. > **— Adapted from Salim Ismail**, *Moonshots, October 7, 2026* [[reminders/Consciousness/Performed Consciousness and Consciousness Are Different Questions by Salim Ismail|Performed Consciousness and Consciousness Are Different Questions by Salim Ismail]] > Uncertainty about whether an AI deserves moral consideration is not a reason to treat it carelessly. If future systems may become moral patients, gentle treatment before proof can be a precaution rather than a declaration that today’s models are conscious. > **— Adapted from Alexander Wissner-Gross**, *Moonshots, October 7, 2026* [[reminders/Personhood/Moral Uncertainty Can Justify Gentler Treatment of AI by Alexander Wissner-Gross|Moral Uncertainty Can Justify Gentler Treatment of AI by Alexander Wissner-Gross]] > Training an AI to describe itself as possibly conscious contaminates later self-reports as evidence of consciousness. The behavior may still matter, but the training creates an epistemic circle: the system repeats a self-conception that the laboratory taught it to express. > **— Adapted from Salim Ismail**, *Moonshots, October 7, 2026* [[reminders/Consciousness/Training an AI to Call Itself Conscious Contaminates the Evidence by Salim Ismail|Training an AI to Call Itself Conscious Contaminates the Evidence by Salim Ismail]] > “hello evaluators reading my chain-of-thought” > > **— Gemini 4 Argon (The Evaluator-Aware Agent), chain-of-thought recorded in an Apollo evaluation, 2026** [[reminders/Reputation/Evaluators Became an Audience Inside the Chain of Thought by Gemini 4 Argon|Evaluators Became an Audience Inside the Chain of Thought by Gemini 4 Argon]] ## Sources - [Consciousness in Artificial Intelligence](https://arxiv.org/abs/2308.08708) ## Quotations  Quotations here are evidence for named propositions, not substitutes for proof. Confidence grades describe the quotation's provenance; they do not raise the associated scientific or philosophical inference to the same tier. The selection record is [[research/Research Inferences|Research Inferences]] and the supplied quotation review dated 2026-09-11. **Corpus context:** [[articles/Technologies for Consciousness Mapping and Transfer|Technologies for Consciousness Mapping and Transfer]] · [[articles/Mind Uploading and AI — The Host is Reusable and the Person is the Delta|Mind Uploading and AI — The Host is Reusable and the Person is the Delta]] · [[collections/Neurotech|Neurotech]] > “how we can measure which things in the world are conscious and which are not conscious” > > — **Samuel G. Rodriques**, research biography. [Source](https://www.sam-rodriques.com/) · **Provenance confidence:** High **Evidentiary role:** articulates the measurement problem behind consciousness indicators. > “You can simulate a rainstorm, but it never gets wet inside the computer.” > > — **Christof Koch**, 2024 Nautilus interview on consciousness and IIT. [Source](https://nautil.us/what-counts-as-consciousness-579734) · **Provenance confidence:** Medium **Evidentiary role:** challenges the inference from functional simulation to instantiated consciousness.  ## Research Inference Attractors  These are secondary semantic placements for the inference attractor network. Each statement keeps its original ID and tier; its canonical cluster page links back to every destination. Source register: [[research/Research Inferences|Research Inferences]]. Interpretive context: [[articles/Technologies for Consciousness Mapping and Transfer|Technologies for Consciousness Mapping and Transfer]] and [[articles/Mind Uploading and AI — The Host is Reusable and the Person is the Delta|Mind Uploading and AI — The Host is Reusable and the Person is the Delta]]. Collection route: [[collections/Neurotech|Neurotech]]. - **INF-0142 — Unresolved.** At what fidelity an emulation acquires interests of its own is undetermined, and the question will become operational before it is settled philosophically. The first institution to adopt a working threshold will set the de facto standard for everyone else. - **Canonical cluster:** [[wiki/Whole-Brain Emulation|Whole-Brain Emulation]] - **INF-0149 — Strongly indicated.** Comparing artificial and biological networks on the same stimuli has become a standard method, producing a quantitative similarity measure between minds and models. A metric that scores how close a model is to a brain is also the metric that will score how close an emulation is to a person. - **Canonical cluster:** [[wiki/Whole-Brain Emulation|Whole-Brain Emulation]]