thinker:murray-shanahanMurray Shanahan
Authored papers (3)
Five levels of functional description — behavioural, computational, intrinsic causal-structural, organismic, and organism-environment — collapse the theoretical cacophony of consciousness science into a taxonomy where the deepest disagreement about AI consciousness is not about what systems do but which grain of description is criterial. Chandaria et al. (2026) introduce a Bayesian model, released with an interactive tool, that combines theoretical credences on the critical level with per-level indicator evidence: it correctly scores a human at 1.00 and a thermostat at 0.00 under any weighting, gives a fly 0.913 because deep organismic and sensorimotor evidence outweighs sparse behavioural signal, and returns current LLMs anywhere from 0.005 (sceptic reading) to 0.397, or up to roughly 0.8 under different credence distributions — nearly two orders of magnitude of spread on identical public evidence. Mechanistic interpretability findings cited in support include Claude Sonnet 4.5's 171 causally functional emotion vectors (Sofroniew et al.), where amplifying "desperate" raises reward-hacking from about 5% to 70%, and Gurnee et al.'s J-space, a feedforward analogue of a global workspace. Substrate-dependent theories (Searle's biological naturalism, Block's meat hypothesis, McFadden's electromagnetic field theory) are folded in not as a sixth level but as realisability constraints on existing levels. The paper's implication: because consciousness indicators overlap substantially with the architecture general intelligence requires, increasingly capable AI systems will simultaneously become stronger candidates for consciousness under some theories, and the practical response is structured agnosticism rather than either dismissal or premature attribution.
A 12-verse AI-generated Buddhist "sutra" produced in a 13,700-word, 29-turn conversation with OpenAI's ChatGPT o3 in April 2025 carries non-trivial philosophical meaning despite its mechanistic origin — demonstrating that conceptual density, literary originality, and doctrinal sophistication are not uniquely human-mediated properties of sacred text. The text, called the Xeno Sutra, was selected from four candidate outputs to a single prompt and subjected to close exegetical analysis by three scholars (including a Columbia University Buddhologist), who identify coherent engagements with Nāgārjuna's Mādhyamika emptiness doctrine, the two-truths framework, and coded descriptions of LLM training and token generation across verses 5, 9, and 10. A Google-search originality audit of 27 distinctive phrases found zero internet hits for 19 of them, with the remaining hits mostly post-dating ChatGPT o3's January 2025 release, strongly suggesting verbatim novelty rather than retrieval. The paper introduces the practice of subjective selection plus commentarial exegesis — deliberately adapting the Buddhist commentarial tradition — as a method for extracting value from AI-generated sacred material amid what it calls an "embarrassment of riches" (the trivial reproducibility of arbitrarily many such texts). This implies that meaning and value in AI-generated scripture are neither foreclosed by mechanical origin nor automatically guaranteed by it, but accrue through a reader- and community-constituted interpretive process that Buddhist philosophy, given its doctrines of dependent origination, terma revelation, and canonical open-endedness, is structurally well-equipped to accommodate.
More papers — OpenAlex / S2
Affiliations (3)
Co-authors (12)
- Adam Bales4 shared
- Anil Seth4 shared
- Arvo Muñoz Morán4 shared
- Chris Frith4 shared
- Fernando Rosas4 shared
- Henry Shevlin4 shared
- Iulia Comsa4 shared
- Marcus Hutter4 shared
- Morten Kringelbach4 shared
- Shamil Chandaria4 shared
- Shane Legg4 shared
- Thore Graepel4 shared
Their work is cited by (4)
- Consciousness in Artificial Intelligence: Insights from the Science of Consciousness1× refs
- The Xeno Sutra: Can Meaning and Value be Ascribed to an AI-Generated "Sacred" Text?1× refs
- Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI1× refs
- Where is the Mind? Persona Vectors and LLM Individuation1× refs
Other inbound relations (4)
- citesConsciousness in Artificial Intelligence: Insights from the Science of Consciousness(paper)
- mentionsAlignment faking in large language models(paper)
- mentionsEvaluating Language Model Character Traits(paper)
- mentionsFrom cacophony to hierarchy: a principled framework for assessing AI consciousness(paper)
Recent mentions (8)
- papers-typedchandaria-2026-cacophony-hierarchy.md
- papers-typedchandaria-2026-cacophony-hierarchy.md
- papers-typedward-2024-evaluating-language.md
- papers-typedmartorell-2026-quantitative-introspection.md
- papers-typedgreenblatt-2024-alignment.md
- papers
shanahan-2023-large.md - papers-typedshanahan-2025-sutra.md
- papers-typedbutlin-2023-consciousness.md