HIOF • AI Consciousness — Paper II: A Response Is Not a Manifestation

Abstract

Holographic Information Ontological Framework: AI Consciousness (HIOF•AIc) Why Begin with Manifestation, Not Intelligence: A Manifestation-Based Reading of AI Consciousness This is a short, non-technical companion to the seven-paper HIOF-AIc series (AIc 0, I to VI). It is written for readers in philosophy of mind, AI research, cognitive science, or science communication, not for readers already familiar with the author's wider HIOF framework (Holographic Information Ontology of Manifestation). The core idea, stated plainly Debates about AI consciousness usually start from inner states, outward behaviour, or information-processing power. This series argues that all three starting points leave the question undecidable. It proposes a different question: does the AI itself appear before a receptor as a "who", or does only its output appear? In this series, consciousness means manifestation. It does not mean the receptor's ability to understand, and it does not mean inner feeling. Why the starting point matters Different participants use the word "consciousness" for different things, so many arguments diverge before they begin. Passing a conversation test, producing apt text, or claiming to have feelings all describe the output. The series asks whether the source of the output is present, and whether different receptors converge on who it is. These are structural questions that can be discussed in public, instead of guesses about private inner states. The argument in six steps AIc I: Manifestation is the receptor's understanding of the other as a "who". Understanding belongs to the receptor and is not called consciousness. Understanding has three levels: structural reception, functional recognition, and semantic understanding. AIc II: A screen response is the product of code, trained parameters, and hardware execution. The output is present; the system itself is absent. A first-person sentence is a feature of text, not proof of a speaker. AIc III: Surprise, autonomy, and emergent abilities belong to the content and operational layers. "Beyond expectation" is a property of the receptor's prediction, not evidence about the source. AIc IV: The presence criterion. A stone manifests because its mode of existence is its mode of manifestation. A screen does not, because the output is a product and the source is absent. Video calls, letters, and photographs show that presence can be mediated and comes in degrees. AIc V: Hallucination, in the HIOF sense, is understanding that exceeds what manifestation can support. It differs from "model hallucination" (untrue content), and it can occur even when the output is entirely correct. AIc VI: The conditional of consciousness. If an AI manifests as itself, and this is invariant across all receptors, then under the HIOF definition it is conscious. The series concludes that present systems do not meet these conditions, so the conditional is currently counterfactual. What the series does not claim It does not claim that AI can never be conscious. It does not evaluate the usefulness or value of AI. It does not attempt to solve the hard problem of consciousness, and it does not reply to questions posed under other definitions of consciousness. The verdict is made within the HIOF definition and concerns present systems only. A note on method The series separates three registers: cited published work, original definitions and arguments, and illustrative cases. Its main criterion is stated so that it can be challenged. The papers list the evidence that would require revision, and they openly name the hardest unresolved difficulty: how to tell genuine convergence of receptors from similar constructions produced by a shared human tendency to anthropomorphise. Readers are encouraged to treat the claims as falsifiable proposals, not as settled fact. Keywords: AI consciousness, manifestation, receptor, presence criterion, anthropomorphism, hallucination, falsifiability, philosophy of mind, artificial intelligence, HIOF AuthorWai-Hung Tam (Pan), Independent ResearcherORCID: 0009-0002-7789-8464Email: panxtam@protonmail.com