Organoid Intelligence - Bryant McGill

# Organoid Intelligence **Entity class:** Concept or analytic term **Domain:** Biocomputing / Brain Organoids / Biohybrid Systems **Doc Type:** Concept and Field Node **Maturity:** Experimental research field with early commercial platforms **Primary Source:** [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]] ## Definition **Organoid intelligence** is the use of self-organized neural tissue—especially brain organoids derived from induced pluripotent stem cells—as an adaptive computational substrate. Machine learning, microelectrode arrays, stimulation systems, and trained readouts form the interface around the living network. The field is distinct from [[wiki/Neuromorphic Computing|neuromorphic computing]]. Neuromorphic hardware reproduces selected neural principles in engineered devices; organoid intelligence uses living neurons and their intrinsic plasticity, metabolism, and developmental organization. ## Platform map - **DishBrain** demonstrated task-structured adaptation by coupling cultured neurons to a closed-loop Pong environment. - [[wiki/Cortical Labs|Cortical Labs]] develops the **CL1**, combining living neurons, silicon interfaces, life support, embedded compute, the biOS software environment, and remote access through Cortical Cloud. - [[wiki/FinalSpark|FinalSpark]] operates the cloud-accessible **Neuroplatform**, placing human-neuron organoids on multielectrode arrays and exposing them through a programming interface. - **Brainoware** uses a brain organoid as the dynamical reservoir in a reservoir-computing architecture, with a trained readout extracting task-relevant behavior. ## Technical constraints Scaling requires longer-lived tissue, vascularization or perfusion, reproducible differentiation, higher interface bandwidth, stable calibration, and explicit provenance for substrate source and experimental history. A living network's adaptive response is evidence of computation in a bounded task; it is not evidence of personhood, unconstrained intelligence, or consciousness transfer. ## Governance boundary The field raises questions that ordinary hardware does not: the experiential status of increasingly complex tissue, the rights attached to patient-derived cells, acceptable stimulation regimes, and the point at which biological research protections become inadequate. Those questions remain open and should travel with the technical record rather than be appended after deployment. ## Relationships **Related cluster nodes:** [[collections/Neurotech|Neurotech]] · [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] · [[wiki/Closed-Loop BCI|Closed-Loop BCI]] · [[wiki/Neural Data Provenance|Neural Data Provenance]]  This entry is routed through [[collections/Neurotech|Neurotech]] and [[collections/Consciousness Continuity|Consciousness Continuity]]. Its source context is developed in [[articles/2026 Annual Report on Brain-Computer Interfaces|2026 Annual Report on Brain-Computer Interfaces]], [[articles/Technologies for Consciousness Mapping and Transfer|Technologies for Consciousness Mapping and Transfer]], [[articles/The Architecture of Continuity and Emerging Neuroinformatics Standards|The Architecture of Continuity and Emerging Neuroinformatics Standards]], and [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]]. Status-qualified source edges are preserved in the terminal Research Edges section. ### Technology and research relationships - [[wiki/BCI ecology|BCI ecology]] structurally integrates **organoid intelligence**. Cross-article architecture. ### People and institutional relationships - **organoid intelligence** is associated with [[wiki/Thomas Hartung|Thomas Hartung]]. Johns Hopkins framing/roadmap. - **organoid intelligence** is associated with [[wiki/Lena Smirnova|Lena Smirnova]]. ### Key Relationships - **organoid intelligence** is identified with the institution [[wiki/Johns Hopkins Bloomberg School of Public Health|Johns Hopkins Bloomberg School of Public Health]]. - **organoid intelligence** is associated with [[wiki/drug testing and personalized neurobiology|drug testing and personalized neurobiology]]. ### Additional Documented Relationships - [[wiki/Neurotechnology Ecosystem|Neurotechnology Ecosystem]] structurally integrates **Organoid Intelligence**. Cross-article architecture. - [[wiki/Brain-Computer Interfaces|Brain-Computer Interfaces]] structurally integrates **Organoid Intelligence**. Cross-article architecture.  ## Neurotech cluster route **Collection:** [[collections/Neurotech|Neurotech]] **Source articles:** [[articles/2026 Annual Report on Brain-Computer Interfaces|2026 Annual Report on Brain-Computer Interfaces]] · [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]] ## Related Work in the Corpus  - In [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]], **Movement II — The Organic Substrate: Living Neural Tissue as Computation** provides the narrative context for **Organoid Intelligence**: Wiki route: organoid intelligence · Cortical Labs and the CL1 · FinalSpark Neuroplatform · biohybrid neural systems · closed-loop interfaces. - In [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]], **Movement VIII — Hybrid and Biohybrid Architectures: The Convergence of Substrates** provides the narrative context for **Organoid Intelligence**: Wiki route: biohybrid neural systems · organoid intelligence · neuromorphic computing · closed-loop coupling.  ## Research Edges  #### Master relationship graph patch — 2026-09-10 **Resolved aliases:** `organoid intelligence` **Collection:** [[collections/Neurotech|Neurotech]] **Relationship source:** [[research/Neurotechnology Ecosystem Relationship Graph - 2026-09-10|Neurotechnology Ecosystem Relationship Graph — 2026-09-10]] **Related source articles:** [[articles/The Architecture of Continuity and Emerging Neuroinformatics Standards|The Architecture of Continuity and Emerging Neuroinformatics Standards]] · [[articles/2026 Annual Report on Brain-Computer Interfaces|2026 Annual Report on Brain-Computer Interfaces]] · [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]] #### Outgoing typed edges - **Edge 324 — `field_leader` → [[wiki/Thomas Hartung|Thomas Hartung]]** — **VERIFIED**; evidence `OI_JHU`. Johns Hopkins framing/roadmap. - **Edge 325 — `field_leader` → [[wiki/Lena Smirnova|Lena Smirnova]]** — **VERIFIED**; evidence `OI_JHU`. Johns Hopkins co-lead and ethics/research bridge. - **Edge 326 — `institution` → [[wiki/Johns Hopkins Bloomberg School of Public Health|Johns Hopkins Bloomberg School of Public Health]]** — **VERIFIED**; evidence `OI_JHU`. - **Edge 327 — `conceptual_bridge` → drug testing and personalized neurobiology** — **VERIFIED**; evidence `OI_JHU`, `CORTICAL_CL1`. Same tools support computation and disease/compound-response research. #### Incoming typed edges - **Edge 3 — [[wiki/Neurotechnology Ecosystem|BCI ecology]] `structurally_integrates` → this entry** — **CORPUS**; evidence `CORPUS_BCI`, `CORPUS_ATLAS`, `CORPUS_CONT`. Cross-article architecture.   #### Consciousness mapping and transfer graph patch — 2026-09-10 **Resolved aliases:** `organoid intelligence` **Collections:** [[collections/Neurotech|Neurotech]] · [[collections/Consciousness Continuity|Consciousness Continuity]] **Relationship source:** [[research/Consciousness Mapping and Transfer Ecosystem Relationship Graph - 2026-09-10|Consciousness Mapping and Transfer Ecosystem Relationship Graph — 2026-09-10]] **Related source articles:** [[articles/Technologies for Consciousness Mapping and Transfer|Technologies for Consciousness Mapping and Transfer]] · [[articles/The Architecture of Continuity and Emerging Neuroinformatics Standards|The Architecture of Continuity and Emerging Neuroinformatics Standards]] · [[articles/2026 Annual Report on Brain-Computer Interfaces|2026 Annual Report on Brain-Computer Interfaces]] · [[articles/The Organic-Synthetic Brain Atlas|The Organic-Synthetic Brain Atlas]] #### Outgoing typed edges - **Edge 324 — `field_leader` → [[wiki/Thomas Hartung|Thomas Hartung]]** — **PRIOR_NEUROTECH_VERIFIED**; evidence `OI_JHU`. Johns Hopkins framing/roadmap. - **Edge 325 — `field_leader` → [[wiki/Lena Smirnova|Lena Smirnova]]** — **PRIOR_NEUROTECH_VERIFIED**; evidence `OI_JHU`. Johns Hopkins co-lead and ethics/research bridge. - **Edge 326 — `institution` → [[wiki/Johns Hopkins Bloomberg School of Public Health|Johns Hopkins Bloomberg School of Public Health]]** — **PRIOR_NEUROTECH_VERIFIED**; evidence `OI_JHU`. - **Edge 327 — `conceptual_bridge` → [[wiki/drug testing and personalized neurobiology|drug testing and personalized neurobiology]]** — **PRIOR_NEUROTECH_VERIFIED**; evidence `OI_JHU`, `CORTICAL_CL1`. Same tools support computation and disease/compound-response research. #### Incoming typed edges - **Edge 3 — [[wiki/Brain-Computer Interfaces|BCI ecology]] `structurally_integrates` → this entry** — **PRIOR_NEUROTECH_CORPUS**; evidence `CORPUS_BCI`, `CORPUS_ATLAS`, `CORPUS_CONT`. Cross-article architecture.  ## Research Inferences  These entries translate the forward-looking register in [[research/Research Inferences|Research Inferences]] into ordinary wiki prose. The tier labels apply to the inference, not automatically to every factual anchor inside it. The interpretive frame comes from [[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]]. ### Biological compute and wetware - **INF-0151 — Established.** A 20-unit CL1 rack at NUS, roughly 16 million living human neurons under datacenter operations with DayOne infrastructure, is the first independently operated biologically integrated server rack. Wetware has entered the rack form factor, with power, cooling, and an operator. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/FinalSpark|FinalSpark]] - **INF-0152 — Established.** DishBrain's roughly 800,000 neurons on a CMOS array learning Pong, and the same substrate class later learning Doom, established that cultured tissue performs closed-loop goal-directed adaptation. The lineage from that demonstration to a commercial rack took four years. - **Attractor routes:** [[wiki/DishBrain|DishBrain]] · [[wiki/Brett Kagan|Brett Kagan]] - **INF-0153 — Established.** CL1 neurons survive up to six months under internal life support, which sets the current replacement cycle for biological compute. Every business model in the sector is implicitly a bet on extending that window, and extension is a tissue-engineering result rather than a computing one. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/FinalSpark|FinalSpark]] - **INF-0154 — Strongly indicated.** A hyperscale developer with roughly 2.1GW of bookings taking a position in biological compute means the substrate is being evaluated against datacenter power economics. That comparison — joules per useful operation — is the only metric under which wetware wins, and it is now being measured by people who buy power. - **Attractor routes:** [[wiki/DayOne|DayOne]] · [[wiki/Biological Data Centre|Biological Data Centre]] - **INF-0155 — Analytic.** A CL1 unit reportedly draws less power than a handheld calculator while hosting millions of neurons, which is a several-order-of-magnitude efficiency claim against silicon inference. If it survives independent measurement, biological substrate becomes economically rational for specific workload classes rather than merely interesting. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] - **INF-0156 — Plausible.** The claimed advantage of biological substrate is learning from sparse data and adapting under changing conditions, which is precisely where gradient-trained networks are weakest. Complementarity rather than replacement is the realistic deployment shape, and complementary systems become permanent. - **Attractor routes:** [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] · [[wiki/Cortical Labs|Cortical Labs]] - **INF-0157 — Established.** CL1 neurons are reprogrammed from human blood cells, which means any donor can supply the substrate. Donor-specific biological compute is therefore available today as a technical matter, and personalized neural substrate is an ordering question rather than a research question. - **Attractor routes:** [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] · [[wiki/Cortical Labs|Cortical Labs]] - **INF-0158 — Plausible.** Growing compute from a specific person's cells creates a substrate with that person's genome running unrelated workloads. The custody, consent, and inheritance questions this raises are unlike anything in either computing or medicine, and they arrive with the first commercial order. - **Attractor routes:** [[wiki/living-cell neural interfaces|living-cell neural interfaces]] · [[wiki/Cortical Labs|Cortical Labs]] - **INF-0159 — Established.** Remote access to organoid electrophysiology from Vevey lets research groups worldwide run experiments on living tissue they never touch. Biological compute as a hosted service preceded biological compute as a product, which mirrors how cloud preceded on-premises consolidation in reverse. - **Attractor routes:** [[wiki/FinalSpark|FinalSpark]] · [[wiki/Neuroplatform|Neuroplatform]] - **INF-0160 — Analytic.** Selling access rather than units means the operator retains the tissue, the protocols, and all the operational knowledge about keeping it alive. Operational knowledge about substrate maintenance is the moat in this sector, not the biology itself. - **Attractor routes:** [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] · [[wiki/FinalSpark|FinalSpark]] - **INF-0161 — Established.** Organoid reservoir computing uses the tissue's intrinsic dynamics as a fixed nonlinear expansion with only the readout trained. It is the cheapest way to extract computation from living tissue, and cheap extraction is what makes early commercialization possible. - **Attractor routes:** [[wiki/Brainoware|Brainoware]] · [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] - **INF-0162 — Strongly indicated.** The field acquired a name, a roadmap, and dedicated funding before it acquired a killer application, which is the normal sequence for capability domains that later become infrastructure. Named fields attract standardized methods, and standardized methods attract industrial participation. - **Attractor routes:** [[wiki/organoid and biohybrid neural systems|organoid and biohybrid neural systems]] · [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] - **INF-0163 — Established.** High-density CMOS microelectrode arrays are the shared instrument across DishBrain, organoid platforms, and slice physiology. One instrument class underlies the entire biological-compute sector, and its channel density improvements propagate to every participant simultaneously. - **Attractor routes:** [[wiki/Microelectrode Array|Microelectrode Array]] · [[wiki/DishBrain|DishBrain]] - **INF-0164 — Strongly indicated.** Organoids stop growing when diffusion cannot supply the interior, so vascularization is the gate on tissue volume. Every claim about scaling biological compute is a claim about perfusion engineering, and perfusion is a solved problem in other organs. - **Attractor routes:** [[wiki/organoid and biohybrid neural systems|organoid and biohybrid neural systems]] · [[wiki/Brainoware|Brainoware]] - **INF-0165 — Established.** A university neurobiology programme maintaining the living cells inside a commercial deployment makes the academic laboratory an operational dependency of a datacenter. That is a genuinely new institutional relationship, and it will be replicated wherever biological racks are installed. - **Attractor routes:** [[wiki/Rickie Patani|Rickie Patani]] · [[wiki/living-cell neural interfaces|living-cell neural interfaces]] - **INF-0166 — Unresolved.** Cortical Labs' stated position distinguishes responsiveness and learning from consciousness, and no measurement adjudicates that distinction. The threshold question will be forced by scale rather than by philosophy: a thousand-unit deployment invites the question in a way a benchtop culture does not. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/FinalSpark|FinalSpark]] - **INF-0167 — Plausible.** Drug discovery, humanoid robotics, cybersecurity, and fraud detection are the named target domains, and only the first has an obvious biological rationale. The others suggest the pitch is energy efficiency and adaptive learning, meaning wetware is being sold as an inference substrate rather than as a biology tool. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/Neuroplatform|Neuroplatform]] - **INF-0168 — Established.** A CL1-based initiative announced by an Italian IT firm with a university partner in January 2026 shows the platform diffusing into ordinary enterprise IT contexts. Diffusion into enterprise is how a novel substrate acquires the tooling and integrations that make it durable. - **Attractor routes:** [[wiki/Brainoware|Brainoware]] · [[wiki/FinalSpark|FinalSpark]] - **INF-0169 — Plausible.** A rack containing both silicon accelerators and living tissue under one orchestration layer is a heterogeneous compute problem of a familiar type. Schedulers already place workloads across CPU, GPU, and specialized silicon; adding a biological device class is an extension, not a revolution. - **Attractor routes:** [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] · [[wiki/organoid and biohybrid neural systems|organoid and biohybrid neural systems]] - **INF-0170 — Strongly indicated.** An operating system that presents living neurons through a code-deployable interface abstracts biology into an API. Once biology has an API, software engineers rather than neuroscientists become the majority of the people programming tissue, and the field's growth rate changes accordingly. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/living-cell neural interfaces|living-cell neural interfaces]] - **INF-0171 — Analytic.** Biological compute is being positioned explicitly against the energy intensity of AI datacenters, which is the most politically durable argument available in jurisdictions with grid constraints. Energy policy will therefore shape wetware adoption more than neuroscience will. - **Attractor routes:** [[wiki/Sustainability|Sustainability]] · [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] - **INF-0172 — Established.** Cell-line suppliers sit upstream of every biological computer, making reprogramming companies a chokepoint in the sector. Supply-chain analysis of biological compute looks like semiconductor analysis with different vendors and the same structure. - **Attractor routes:** [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] · [[wiki/living-cell neural interfaces|living-cell neural interfaces]] - **INF-0173 — Plausible.** If a person's residual state can run on any sufficient substrate, living tissue grown from that person's own cells is one candidate host among several. The biological option reframes uploading as substrate choice rather than as escape from biology. - **Attractor routes:** [[wiki/Synthetic Biological Intelligence|Synthetic Biological Intelligence]] · [[wiki/living-cell neural interfaces|living-cell neural interfaces]] - **INF-0174 — Analytic.** A six-month substrate lifetime forces periodic migration of whatever state the tissue holds, which means biological compute must solve state portability immediately rather than eventually. The shortest-lived substrate will produce the best migration tooling. - **Attractor routes:** [[wiki/Cortical Labs|Cortical Labs]] · [[wiki/Biohybrid Neural Systems|Biohybrid Neural Systems]] - **INF-0175 — Unresolved.** Whether a rack of human neurons is a medical device, a research material, a computing product, or a biological specimen is unsettled in every jurisdiction where one has been installed. Classification will be determined by the first customs declaration or export-control question, not by legislation. - **Attractor routes:** [[wiki/Neuroplatform|Neuroplatform]] · [[wiki/FinalSpark|FinalSpark]]  ## 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-0485 — Plausible.** A demonstrated brain-to-brain channel exceeding natural language bandwidth between two humans is the marker that coupling has exceeded speech. Any bandwidth above roughly forty bits per second of semantic content qualifies. - **Canonical cluster:** [[wiki/State Sufficiency Problem|State Sufficiency Problem]] - **INF-0495 — Analytic.** The largest deployment of brain-computer interface technology is a movement-disorder therapy that almost nobody counts as a BCI. Category framing determines what is noticed, and the most consequential deployments are routinely filed under other names. - **Canonical cluster:** [[wiki/State Sufficiency Problem|State Sufficiency Problem]]  ## Simple Reminders, Quotations, and Thoughts > "If the right combination of chemicals can perform thinking and feeling emotions, and it does—the proof being ourselves—then sure there should be many other analogous mechanisms for doing the same." > **— Carlo Rovelli**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Consciousness/Chemistry Proves Minds May Have Other Mechanisms by Carlo Rovelli|Chemistry Proves Minds May Have Other Mechanisms by Carlo Rovelli]] > "One gloomy possibility is that we become zombie consumers of a machine-run world straight out of an apocalyptic futuristic film noir." > **— Athena Vouloumanos**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Humans Could Become Consumers of a Machine-Run World by Athena Vouloumanos|Humans Could Become Consumers of a Machine-Run World by Athena Vouloumanos]] > "The patterns involved can easily exceed what the human mind can grasp." > **— Bart Kosko**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Machine Patterns Can Exceed Human Understanding by Bart Kosko|Machine Patterns Can Exceed Human Understanding by Bart Kosko]] > "In fact, natural cognition is likely much more complex and detailed than our current incarnations of artificial intelligence or cognitive computing." > **— Maximilian Schich**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Natural Cognition Is More Complex Than Today’s AI by Maximilian Schich|Natural Cognition Is More Complex Than Today’s AI by Maximilian Schich]] > "One intriguing possibility is that for a machine to think about thinking, it will need to have something like free will." > **— Hans Halvorson**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Thinking About Thinking May Require Free Will by Hans Halvorson|Thinking About Thinking May Require Free Will by Hans Halvorson]]