The Bee in the Rack: Human Neural Cultures, Commercial Computing, and the Limits of the Size Argument

Abstract

Living human neurons, grown from induced pluripotent stem cells on multielectrode arrays, are now rented as a commercial computing service. Cortical Labs operates racks of its CL1 units in Melbourne and, with partners, in Singapore. That moves the debate over artificial consciousness onto ground where its most developed skeptical argument, that silicon is the wrong substrate, cannot be used in its usual form. In an October 2026 broadcast, the company's chief scientific officer said there is no evidence that small groups of cells in a dish have feelings and that the company checks constantly, and a neuroscientist from Imperial College London, interviewed for the same segment, said that the cultures will not become conscious and that computing on a biological substrate is no different from simulating a large artificial intelligence system, so that the same ethical principles apply. Read as resting on neuron count, the company's reasoning fails against its own comparison class: a single unit holds cells on the order of a fruit fly brain's neurons, and the leading account of organoid sentience rejects size as a reason for dismissal. Read as resting on simple organization, it holds for present flat cultures, and only while they stay simple. Human thresholds that might seem to supply a floor are metabolic rates and cannot be converted into neuron counts. On either reading, the prediction that the cultures will not become conscious gives no assurance about the cultures the research program is engineering. The equivalence claim faces a dilemma. If biology matters to consciousness, the skeptic cannot dismiss these cultures with the arguments about material used against silicon and is left with a contested claim about organisms that neither speaker invoked. If it does not, reassurance must rest on an organizational claim no one on camera made, while the operator's research changes that organization. The operator's own laboratory has disclaimed in print the equivalence of status on which the reassurance would depend. The equivalence also omits the people whose cells and genomes are in the dish. The research lines trace to neonatal foreskin tissue distributed for laboratory research, no consent documentation for the commercial product is public, and donor protections written for research may not reach it. The paper makes no finding that the cultures are conscious and proposes no prohibition. It sets out review triggers keyed to organization, one already met by the operator's research, alongside documentation of training signals, provenance disclosure, decommissioning records, measurement research, and regulatory clarification.