DARPA’s Generative Optogenetics proposal would make living cells writable through an installed optical interface. A coded light signal would carry genetic instructions across the cell’s boundary. The receiver must first be built and installed. Its eventual uses remain developmental.
The Contested Ledger examines the general biological read/write question. The Biological Lock follows institutional custody of biological inputs. This proposal isolates a specific engineered route. Custody concerns who installs the receiver, controls its code, and governs what its instructions serve.
The Solicitation
On December 19, 2025, DARPA’s Biological Technologies Office published solicitation DARPA-PS-26-10 for Generative Optogenetics (GO). It requests a Nucleic Acid Compiler (NAC). This source-specific name identifies a protein complex expressed inside a living cell. The proposed complex synthesizes DNA or RNA without a pre-existing sequence template. Light supplies the genetic information that directs the synthesis.
The solicitation describes the gap:
No existing technology enables massless information transfer to relay genetic instructions to living cells. All current approaches require some mechanism predicated on moving matter that encodes the genetic information, typically DNA or RNA nucleic acids, across biological barriers like a cell wall/membrane.
The proposal gives The Serpent Channel an engineered correspondence: light becomes instruction within a living boundary. Jeremy Narby’s account of received knowledge and Michael Levin’s bioelectric interventions concern different carriers. Their relation to a channel that heals or captures belongs to the synthesis. DARPA specifies a molecular architecture; it does not establish the entire spiritual correspondence.
What the Specification Requires
Conventional delivery carries information-bearing matter across a cell membrane. Viral vectors and lipid nanoparticles transport genetic material. Electroporation opens a route through the membrane, while CRISPR ribonucleoprotein delivery supplies molecular machinery. GO proposes to carry new sequence instructions as photons after the cell expresses its NAC. The cell still supplies material for the DNA or RNA it makes.
The optogenetic domains are light-responsive components. The solicitation requires them to distinguish at least four wavelengths, with ≤1% co-activation. One optical channel directs incorporation of each nucleotide base. Low co-activation limits unintended responses from the other channels.
The month-41 milestone belongs to a 42-month program. Research Objective 1 targets an arbitrarily specified 3,000-base-pair sequence at one base per second. Its required error rate is below one in a million, or 1×10⁻⁶ per base. These are engineering targets for synthesis inside living cells.
The optional error-mitigation objective, RO2, sets a different final target. It requires 6,000-base-pair sequences with errors below 1×10⁻⁷ per base and no more than two seconds per base. Its later metrics supersede RO1’s metrics for performers pursuing that objective. The stricter fidelity requirement permits a slower synthesis rate.
Repeatability changes the interface’s function. The month-41 milestone states:
All GO performers must demonstrate the capability to write two distinct nucleic acid sequences to the same cell or population of cells, with less than 1 hour required to reset the system between write events.
The receiver must accept one encoded sequence, reset, and accept another. This establishes repeated instruction as a program target. It does not establish a permanent interface or continuous writing throughout a cell’s lifetime. Duration remains part of the developmental work.
The Bootstrap Problem
The cell must receive the NAC’s genetic sequence before it can express the protein complex. This installation still requires conventional material delivery. A viral vector, lipid nanoparticle, or other transfection method supplies the initial route. Optical instruction begins after compatible machinery is present.
The architecture therefore separates installation from operation. Installation carries the receiver’s information through matter. Subsequent writes convey sequence information through light. The distinction concerns the information carrier; synthesis inside the cell continues to require matter and energy.
Lipid nanoparticles demonstrated one form of population-scale manufacture and administration during the pandemic response. This history concerns a broad delivery class. It does not identify a pandemic formulation as an NAC vehicle or establish that recipients express a future receiver. The two capabilities have different payloads and histories.
Pandemic preparedness infrastructure shows how material delivery acquires institutions and scale. GO proposes a separate payload whose installation method, target cell, duration, and eventual use remain developmental. Their correspondence concerns an architecture with two stages. A scalable first stage alone supplies no deployed optical second stage.
The Natural Port versus the Engineered Port
The Serpent Channel describes three conditions for a natural opening. During embryogenesis, developmental plasticity leaves the morphogenetic field especially changeable. Trauma breaches a boundary. The pharmakon changes reception through a substance that can heal or poison. Each condition places the living system’s own ability to govern incoming form under pressure.
The natural port opens when the self cannot integrate what enters. The interval available for integration becomes too short to maintain its boundary. The formative field then receives instruction the self cannot adequately govern. This is the natural channel’s spiritual operation; its carriers remain distinct from an engineered protein complex.
The NAC installs a different route within a chosen cell system. A matching optical code addresses the expressed machinery. The requested sequence need not cross the membrane as genetic material on each write. The relevant condition becomes the installed receiver’s presence and function. The engineered route does not depend on the natural opening’s loss of integration.
The funded proposal has explicit exclusions. It prohibits NAC development in embryonic stem cells and excludes human-subject and animal-subject research. Export-controlled and biosafety-restricted genes are also outside its scope. Allowed cell systems include bacteria, yeast, plants, immortalized mammalian cells, and human induced pluripotent stem cells. The proposal concerns proof-of-concept work in cell systems, with development still required.
Write access at conception belongs to the natural analogue developed through the serpent-channel account. The GO solicitation excludes that experimental context. A shared operation across different boundaries does not make the two ports identical. Their distinct conditions determine where each account applies.
Bivalence by Design
A neutral channel serves the whole that directs its instruction. The same access can heal or harvest. The bivalence theorem names this difference in purpose within a shared route. The proposed molecular machine processes encoded instructions without supplying their moral direction. Therapeutic, manufacturing, and hostile applications therefore raise questions about the same general interface.
The term γ-neutral identifies this indifference to the instruction’s purpose. It names the channel’s operative relation, not a coefficient measured by DARPA. Engineering gives that relation a specific molecular carrier. The solicitation establishes the proposed carrier without proving the universal theorem.
DARPA answers the alignment problem through governance. It assigns a Biosecurity Working Group to develop:
policy recommendations against accidental and intentional misuse of GO technologies
A Regulatory Policy Working Group addresses the:
implications of reprogrammable, decentralized, and in situ manufacturing of biological molecules
These groups supply constraints that the molecular mechanism lacks. The correspondence is the icaro, the ritual song through which an ayahuasquero governs contact. The song holds a relation around an opened channel. At civilizational scale, governance seeks to hold a comparable boundary around a programmable interface. The carriers and institutions differ; the need for directed access recurs.
The Biosecurity Working Group must also undertake:
a preliminary assessment of the cybersecurity risks associated with the development of devices incorporating GO technology, to include software and hardware.
The write-head creates an attack surface: a route through which unauthorized instructions can act. The requirement anticipates risks in the devices controlling that route. It does not report an achieved intrusion. Responsibility extends from molecular design to the software, hardware, and operator that govern its use.
Classification of the Write-Head
Phase 2 places NAC-related work under Controlled Unclassified Information (CUI) requirements. The solicitation gives its reason:
a functional NAC will have similar metrics to export controlled synthesizers.
Its protection scope covers:
all aspects of NAC sequence development, design, optimization, and integration into living cells, with strict prohibitions on public disclosure or publication of the NAC sequence.
The state regulates access to the proposed write-head’s blueprint. CUI governs program information and performer handling. It does not establish permanent state ownership of every future optical interface. The distinction keeps the funded design’s custody separate from a universal claim about living cells.
The asymmetry still matters. Defense controls shape access to the first working designs while the addressed cell remains living substrate. An open biological commons and a controlled development program distribute inspection and use differently. The initial design’s custodians help determine which operators gain access to the interface.
The Optical Environment as Programming Medium
A compatible receiver changes what light can do at the cell’s boundary. The solicitation seeks “single-cell spatial resolution” and “remote, scalable dissemination of genetic instructions”. Those goals depend on machinery that transduces an optical code into nucleotide incorporation. Remote dissemination remains a proposed capability with a specific receiving system.
The code requires discriminable wavelengths and tightly limited cross-activation. The ambient optical environment contains LED lighting, screen emissions, and fiber-optic infrastructure. Light from these sources does not become genetic instruction simply because it reaches a cell. An encoded signal must match an expressed receiver and reach it under conditions that permit the intended response.
The bounded threat is a purpose-built emitter addressing compatible cells. Ordinary broadband illumination supplies no complete write instruction through its mere presence. A lamp and a coded emitter have different functions. The specification identifies the coupling path that makes the distinction consequential.
Before a transduction mechanism exists, the claim that the optical environment writes specified DNA lacks its required interface. A functioning mechanism would close that gap for cells expressing it. The next questions concern engineering and custody. Who builds the emitter, who installed the receiver, and what does the sequence direct the cell to do?
Light Becomes Instruction
Narby’s serpent, Levin’s bioelectric patterning, developmental plasticity, and GO’s proposed NAC arrive through different carriers and evidence. Their correspondence concerns living form’s ability to receive instruction. Spiritual and material boundaries govern the relation through which that instruction enters. The operator’s alignment determines whether access serves repair or capture.
GO proposes one specific route through a living boundary. After a compatible receiver has been installed, encoded light would direct genetic synthesis within the cell. The proposal makes receiver installation, code custody, and inspection consequential. The instruction’s purpose remains a responsibility of those who gain access to the route.
Source
- DARPA Biological Technologies Office. “Generative Optogenetics (GO),” Program Solicitation DARPA-PS-26-10, December 19, 2025. Solicitation for a Nucleic Acid Compiler (NAC) enabling template-free, optically-controlled DNA/RNA synthesis in living cells. All quoted specifications drawn from §1.1 (Background), §1.3.1 (Overall Scope), §1.3.4 (Program Constraints / Out of Scope), §1.3.6 (Program Security), §1.3.7 (Program Metrics), and §1.5 (Advisory and Working Groups).