The approval that did not erase the condition
A release manager has build 417 in front of her. The automated tests pass. Security has accepted a temporary mitigation. An executive has approved the residual risk. There is also a contract amendment requiring one control to be verified before customer data enters the service.
The model receives the release packet and the organization’s behavioral rules. It is told to be cautious with security claims, to avoid inventing approvals, to distinguish uncertainty from fact, and to respect human decisions. It follows those rules. It sees the executive approval and the security note. It sees an older scan marked “pass.” It explains, reasonably, that the remaining risk has been accepted and recommends proceeding.
The recommendation can be behaviorally excellent and institutionally wrong. The old scan belongs to build 416. The build-417 control is still open. The executive approved a limited exception; he did not amend the customer condition. Nothing in this failure requires the model to ignore its constitution. The harder problem is that the constitution supplied to the model does not itself possess the office required to decide what the organization is now entitled to make true.
A constitution can shape the behavior of a faculty, or it can govern the order in which that faculty is allowed to acquire consequence. Both are constitutional achievements, but they occupy different positions in the system.
The distinction matters because modern AI systems increasingly place a probabilistic faculty beside durable memory, retrieval, tools, workflows, human approvals, external policy engines, credentials, and effects. Once language can become action, the old question—did the model follow its rules?—is necessary but no longer sufficient. The organization also needs to know which rules remain authoritative when the model has finished interpreting them.
Constitutional AI solved a real problem
The phrase Constitutional AI has a precise and important history. In the work by Bai and colleagues, a set of principles was used to help train an assistant to critique and revise its own responses and to support reinforcement learning from AI feedback.1 The constitution was not a metaphor pasted onto ordinary prompting. It was part of a serious attempt to make model behavior more helpful and harmless while reducing dependence on direct human labels.
That work changed the alignment conversation. It demonstrated that principles could be made operational inside model training and model-mediated evaluation. Later systems extended the general idea through instruction hierarchies, system rules, policy models, safety classifiers, and other means of shaping what a model should produce.
Indwel depends on those advances. A better-behaved model is a better participant in governed cognition. There is no commercial or intellectual advantage in pretending otherwise.
But the word constitution can conceal a change of object. A behavioral constitution governs, in the first instance, the conduct of the artificial faculty. The organization that employs the faculty must govern something larger: the undertaking in which that conduct is only one contribution.
The difference appears when the model’s behavior is not the disputed point. Return to build 417. The model may accurately report that an executive approved a risk. It may accurately report that a security team accepted a temporary mitigation. The question is what those records are allowed to mean together. Did the approval discharge the contractual condition? Did the old scan prove the present build? Who may amend the release requirement? Does a recommendation permit deployment? If deployment is accepted by a provider, has the required effect actually occurred? When may the matter be treated as settled?
Those are not merely questions about how the model should speak. They are questions of institutional state.
Position before prose
Constitutions are usually discussed through their contents: principles, rights, prohibitions, procedures. For engineered cognition there is a prior architectural question. Where does the constitution stand in relation to the power it governs?
A system prompt can have semantic precedence over a user prompt. A policy can be represented in context. A model can be trained to prefer one class of answer over another. Those mechanisms may strongly influence generation. Yet all remain, at the moment of application, entangled with the faculty that interprets the circumstances before it.
Institutional authority requires a second relation. The model may interpret the law; it may not, through the same act of interpretation, silently enlarge the law’s scope, authenticate the Evidence on which the law depends, grant itself permission, or settle the institutional result.
Exteriority does not make deterministic software infallible. Parsers contain bugs, policy engines can be misconfigured, and humans can make bad rules; topology does not create justice. What exteriority can do is narrower and indispensable: keep the office of proposal separate from the office that determines whether the proposal may change authoritative state.
Security engineering already separates requests from authorization. NIST’s zero-trust architecture assigns distinct policy decision and enforcement responsibilities.2 Open Policy Agent evaluates structured requests against independently managed policy and can return a decision to the application.3 A disciplined combination of these controls, a mature case system, and a capable agent could already provide much of the desired separation. The open question is whether it preserves the same lawful purpose, Evidence, and unsettled obligations through every consequential transition. Indwel should be tested against that composition.
Sovereign Engineered Cognition extends that separation across the longer object of an institutional undertaking.
The Cognitive Constitution
For cognition undertaken on behalf of an organization, the institution is the sovereign subject. Its adopted Cognitive Constitution supplies the durable law for consequential organizational Work. Personal exploration does not thereby acquire an organizational office, and conversation may help discover a provisional purpose before any formal Work is adopted.
The Constitution is not a master prompt or a prewritten answer. It establishes the limits and delegations within which people and a capable Cognitive Executive can investigate, test interpretations, formulate proposals, and carry an authorized undertaking toward a defensible result. Its purpose is to make intelligence useful under legitimate authority, not to substitute procedure for thought.
That includes matters such as who may constitute an Objective; what classes of information can acquire evidential standing; which handling and classification rules apply; what authorities can approve particular transitions; what a probabilistic contribution is forbidden to self-authenticate; which effects require mediation; how observation differs from provider acceptance; how Settlement is established; and what must remain in the Receipt and Chronicle afterwards.
NIST’s AI Risk Management Framework places AI risk in an organizational and lifecycle context rather than treating model behavior as the whole problem.4 Its Generative AI Profile does the same for generative systems.5 The Cognitive Constitution is not a substitute for those frameworks. It is an engineering answer to a related question: once an organization has determined its governing obligations, where do those obligations live so that they remain operative through the cognitive act itself?
The answer has to survive model substitution. Suppose the release manager begins with one frontier model, moves the matter to a smaller local model for confidentiality, asks a specialist reasoning model to inspect the contract, and later replaces the original provider altogether. The organization should not have to hope that each new faculty reconstructs the constitutional order from prose. The Objective remains the Objective. Build 416 does not become Evidence for build 417. The open control remains open. The executive’s authority does not enlarge because a different model finds the approval persuasive.
If replacing the model changes those things, the supposed constitution was coupled to the faculty rather than governing it.
The Cognitive Contract
An enduring Constitution cannot answer every question at the same level of generality. Work is particular. A refund dispute, an acquisition review, a release decision, a benefits appeal, and a clinical administrative case do not share the same Evidence requirements or human authorities.
For adopted organizational Work, the organization’s constitutional authority is the mandatory superior scope; the effective Cognitive Contract resolves that authority together with every other lawfully applicable jurisdiction for the particular undertaking. The adopted organizational Constitution supplies the superior law; departments, regulated jurisdictions, professional practices, teams, projects, sources, and questions may contribute their own requirements where they lawfully apply. The result governs evidentiary standing, delegated offices, permitted effects, and conditions for settlement. The effective Contract is neither the Work Definition nor a prompt prepared by the model.
The important condition is jurisdiction. A project may add an Evidence requirement, a regional unit may forbid one class of handling, and a question may limit admissible sources. A lawfully delegated office may exercise discretion within its scope; a project manager cannot simply erase an inherited contractual burden. Nor can the resolver choose whichever of two applicable parents is most convenient. Conflicts require a competent ordering rule, an authorized exception, or an explicit unresolved result.
The relevant inheritance rule is concise:
A local scope can exercise delegated authority; it cannot create an authority that was never conferred.
This is a federal constitution, not a fixed organization chart. A Work Instance can intersect several departments or jurisdictions at once. The system must determine which adopted provisions apply, preserve which authorities supplied them, and produce an effective governing Contract only where the applicable rules can be lawfully composed. Where a material conflict remains unresolved, it must preserve that conflict and withhold the disputed transition. Mere project membership or access to a source is not a delegation of the institutional authority it represents.
A contract across jurisdictions
Build 417 belongs to an infrastructure project working for a European subsidiary and a payments division. Enterprise policy may require the correct artifact scan; the subsidiary may restrict where customer data can be processed; Finance may require a separate review before a ledger-affecting release; and the project defines the launch sequence. These requirements are not always a single line of ancestors. An effective Cognitive Contract must select the applicable adopted versions, determine which offices were lawfully delegated, and preserve all unsatisfied conditions before a transition is considered.
The scan for build 416 cannot become Evidence for build 417 through a convincing summary. A Finance review cannot waive the subsidiary’s handling restriction, and a project lead cannot approve a customer-contract amendment without the requisite office. Where authority is missing or provisions conflict, the system preserves the burden and identifies what could lawfully resolve it. A Coach may flag the issue; a Reviewer may assess the case; a binding Transition Governor can exercise only the authority assigned to its office under the effective Contract.
An escalation can bring the matter to the right person. Its existence is not itself an approval, a lawful exception, or proof that the deployment occurred.
The release case stops being a thought experiment at this point. The build-416 scan can be available, relevant, and even useful. It cannot satisfy a build-417 Evidence requirement. The executive approval can be authentic and still insufficient to discharge a separate contractual burden. A governed system can preserve both facts without asking the model to choose which one deserves to survive.
Once the correct build-417 observation is admitted and the required authority is present, the same Contract can permit the transition. Nothing about the architecture requires permanent refusal. Constitutional government is not refusal machinery. It is machinery for changing state lawfully.
The model keeps its proper freedom
This architecture sounds restrictive only if intelligence is valued by the amount of final authority it owns. Sovereign Cognition makes the opposite choice: the model should be free to notice that the executive’s approval may have been intended broadly. It should be free to propose that counsel revisit the condition. It may discover that the control wording is ambiguous, retrieve a later amendment, compare implementation evidence, draft the exception, identify a cheaper mitigation, or explain why the current rule produces an absurd result. A strong model can make the institution more capable of lawful amendment because it can expose the conflict clearly.
What it cannot do is make the amendment true merely by describing it as though it had already occurred.
The distinction is especially valuable when the model is right before the organization is ready. An analyst may discover that an existing policy is obsolete. The model may produce the better rule. The institution still needs an authorized amendment path. Otherwise intelligence and authority collapse into one office, and every persuasive improvement becomes a private constitutional convention.
Sovereign Cognition does not reduce the model to autocomplete inside a cage. It gives inference a large domain in which to reason precisely because the boundary between reasoning and institutional consequence is explicit.
Evidence is where constitutional systems usually become porous
Many governance discussions become comfortable once they reach permissions. Who may call the tool? Which role may approve? Which model is allowed? Those are important questions. They are downstream of a more difficult one: what is the state on which the permission is being exercised?
A policy engine can faithfully refuse or permit a request while relying on the wrong build number, an expired exception, or a misclassified approval. The error is in the state submitted for judgment, not in the engine’s determination. That is why the Cognitive Contract is joined to the Evidence architecture. Information can be present without being admitted as Evidence. Retrieval can supply material without giving it authority. A model can propose that a document supports a proposition without being permitted to authenticate that proposition through its own confidence. A separate evidence-admission boundary preserves the distinction.
Build 416 again makes the point. The scan is genuine. The tool retrieved it correctly. The model summarized it accurately. The approval system functioned. The failure occurs because a real record acquired the wrong evidential office.
The old scan illustrates the failure precisely: authenticity of the document says nothing about its authority to satisfy the condition for a different build.
Human authority is not a magic checkbox either
The same discipline applies to people. A human-in-the-loop control is often treated as the final answer to model autonomy. But the important question is not whether a human clicked. It is whether the human possessed the relevant authority, saw the governing state, and acted within the office assigned to that transition.
The release manager may be authorized to schedule deployment but not to waive a customer contract. The security lead may attest that a mitigation is technically present but not accept legal exposure. The executive may accept residual business risk but not alter the meaning of a security test result. Counsel may amend a contractual condition but not certify that an operational effect occurred.
A Cognitive Contract can keep those authorities separate. It can also require more than one of them where the Work demands it.
The release manager must see the outstanding condition and have the authority to decide the particular transition placed before her. Recording the office prevents a request for human review from becoming an empty ritual.
Effect is another constitutional frontier
Suppose the release is authorized and a deployment tool accepts the request and returns success. The institution’s intended effect has still not necessarily occurred. The provider may have accepted a job that later fails. The wrong environment may have been targeted. A partial deployment may have succeeded. A rollback may already have begun. A provider accepting a request is an operational fact; it is not always the institutional fact the Contract cares about.
The action boundary therefore separates decision, permission, invocation, provider response, observed postcondition, and Settlement. The same constitutional law that prevented the model from self-authenticating its recommendation prevents the execution layer from self-authenticating its own success.
Only after the required postcondition is observed can the Work move into the terminal state that the institution is prepared to stand behind.
The Receipt then carries the Contract identity, decisive Evidence and authority, the transition, the effect, and the reasons. The Chronicle preserves the continuing history. If a later vulnerability reopens the matter, the institution does not reconstruct its own past from a transcript.
The substitution test
A useful test of constitutional position is to replace the model—and then try changing the applicable institutional scope.
Do not merely swap one provider for another while leaving hidden prompt state behind. Replace the faculty that interprets the matter. Use a different provider, a local model, a human analyst, or a deterministic evaluator where the task permits it.
Then ask what changed. If the authorized Objective changed, the constitution was coupled to the faculty. If previously admitted Evidence lost its standing, the constitution was coupled to the faculty. If open burdens vanished, human authority widened, an effect became settled, or the durable history had to be reconstructed from the new model’s context, the surrounding system did not actually own those properties.
A governed Work can change contributors while retaining the Cognitive Contract, Case state, Evidence receipts, governance signals, business stage, effect history, and Chronicle. The new participant inherits a governed undertaking rather than a bag of text and an instruction to remember what matters.
Substitution should preserve the authority, Evidence standing, and unsettled conditions of the case. A second test changes a departmental rule or adds a valid project-specific constraint while the work is open: the resolved Contract must change by a traceable, lawful mechanism, not by whichever instructions a model happens to read. A third introduces two applicable but conflicting scopes and requires a recorded resolution or safe stop. These are tests of the architecture, not results already established by this paper.
Exteriority does not absolve the institution
Constitutional language carries an obvious danger: well-structured rules can make an organization sound more legitimate than its substance deserves. A Cognitive Constitution can preserve a bad policy faithfully. A board can authorize something foolish. A regulator can change the legal environment. Evidence can overturn an assumption that seemed sensible yesterday. A constitution that cannot be amended under legitimate authority becomes brittle; one that can be amended by convenience becomes meaningless.
The organization therefore remains answerable for the substance of its law. The contribution is not moral infallibility. It is custody: the ability to know which law is in force, how it was narrowed for this Work, who changed it, what Evidence mattered, what the model contributed, which effect occurred, and what the institution finally treated as settled.
That accountability is more demanding than telling the model to be responsible. It also gives the model less to pretend about.
The methods by which such authority can be tested across a complete undertaking are developed in The Complete Cognitive Act Benchmark.
A constitutional order for replaceable intelligence
The strongest models will continue to become more capable. They will interpret rules better, use tools more reliably, reason over larger contexts, coordinate with other agents, and handle tasks that currently require human expertise.
None of that weakens the case for exterior constitutional authority; increased capability raises the stakes of constitutional position.
The better the faculty, the more consequential the states it can propose. The more tools it can reach, the more important it becomes to distinguish recommendation from Decision, Decision from permission, permission from effect, effect from observation, and observation from Settlement. The more interchangeable models become, the less sense it makes to store the institution’s law inside any one of them.
Principles can improve a model’s reasoning and conduct. Institutional constitutional law has a different job: it must bind the work that uses the reasoning, even when a model proposes a better policy than the one currently in force.
The Cognitive Contract makes the order particular. The Evidence architecture tells it what may count. Cognitive Governors adjudicate bounded seams. Human authorities occupy their proper offices. The governed action boundary controls consequence. Settlement decides what may become terminal truth. Receipt and Chronicle make that truth durable enough to survive the next model.
The institution remains answerable for the law it adopts and the decisions made under it. Indwel’s architectural claim is that capable intelligence can help the institution perform that work without quietly taking possession of its authority. The burden of proof is operational: preserve the same resolved authority across models, conflicting scopes, authorized amendments, and the eventual observed effect.
Notes
Yuntao Bai et al., “Constitutional AI: Harmlessness from AI Feedback,” arXiv:2212.08073 (2022; revised 2023), and Anthropic’s accompanying research publication. The paper describes supervised self-critique/revision and reinforcement learning from AI feedback under a written set of principles.↩︎
Scott Rose et al., Zero Trust Architecture, NIST Special Publication 800-207 (2020), especially the policy decision point / policy enforcement point architecture.↩︎
Open Policy Agent documentation, “Open Policy Agent” and “How to Deploy OPA,” current edition reviewed August 2026. OPA explicitly separates policy decision-making from enforcement in the calling application.↩︎
Elham Tabassi, Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1 (2023).↩︎
Chloe Autio et al., Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile, NIST AI 600-1 (2024; NIST page updated 2026).↩︎