Sourced Advocate Composite Persona Architecture

Legitimacy is
not a feature.
It is architecture.

A methodology for building AI deliberation agents whose epistemic authority is grounded in the actual communities their outputs affect. Domain-agnostic. Special education is the first application.

"Capability without legitimate source constituency is not AGI. It is a very sophisticated autocrat."

Mark Peter · IEP Sensei QNaN Protocol Syracuse SPED Council · Research Instrument
The problem
Part I · The Problem

The Kardashev Question

The Kardashev scale was invented as a taxonomy of civilizations based on energy: how much a civilization can harness, and from where. A Type I civilization masters planetary energy. Type II, stellar. Type III, galactic. The scale is useful not because it describes where any civilization actually is, but because of the distinction it draws between capacity and distribution. A civilization that reaches Type I energy capacity while maintaining artificial scarcity for the bottom half hasn't solved civilization. It has scaled it. The abundance is real. The architecture of who it reaches is not.

The same logic applies to intelligence. A system with superhuman reasoning capability that draws its epistemic authority from the statistical average of the training corpus, without grounding in the actual communities whose lives it shapes, is not more intelligent. It is more powerful. Those are not the same thing.

Capability without legitimate source constituency is not AGI. It is a very sophisticated autocrat.

Prior AI architecture research has organized itself around three problems: capability (what can the system do?), alignment (what does it optimize for?), and safety (what can't it do?). These are real problems. Substantial work has been done on all three. The fourth problem, whose epistemic position authorizes the output, has not been treated as a design input. It has been treated as a downstream accountability measure: build the system, deploy it, audit the disparate outcomes, write the correction policy.

That sequence assumes the legitimacy problem can be retrofitted. SACPA begins from the proposition that it cannot. Legitimacy is not a feature you add at the end. It is either a design input or it is absent.

Special education is not an obvious choice if you are thinking about headlines. It is an obvious choice if you are thinking about the problem. The domain has practitioners already deploying AI on high-stakes decisions, including eligibility determinations, placement recommendations, and service allocations, whose outputs directly set a child's educational trajectory. The people most affected are the people most systematically excluded from the rooms where decisions get made: disabled students, families of color, non-speaking children who communicate through augmentative technology.

Special education is also, already, a healthcare domain. School psychologists conduct psychological and neuropsychological evaluations that carry diagnostic weight. Students with Other Health Impairment eligibility have chronic medical conditions such as epilepsy, ADHD, Type 1 diabetes, and complex psychiatric diagnoses, whose management directly shapes every goal in their IEP. Transition planning for students with significant medical needs requires coordination across physicians, therapists, and adult healthcare systems that most schools are not equipped to navigate. The AI legitimacy problem in special education is not separable from the AI legitimacy problem in clinical care. It is the same problem, compressed into a single room.

The disability rights community named the principle at stake decades before AI made it urgent: Nothing About Us Without Us.


Part II · The Architecture

What SACPA Is

SACPA, Sourced Advocate Composite Persona Architecture, is a methodology for building AI deliberation agents in any domain where the difference between what is known, what is inferred, and what is speculated has real consequences for real people. Three properties distinguish SACPA agents from AI systems that merely adopt a professional framing.

Sourced
Knowledge is tied to named practitioners, researchers, and self-advocates, not the diffuse statistical average of a training corpus. When a SACPA composite cites methodology, the claim is traceable to documented positions of named practitioners. The source is not "research suggests." The source is a person, in a context, with a documented view.
Claim-disciplined
A sourced composite distinguishes what it knows from what it infers from what it does not know. It surfaces genuine disagreement within its source constituency rather than averaging the controversy into comfortable consensus. The field is not internally consistent. An honest composite holds that disagreement.
Scoped
A school psychologist composite speaks from within the documented scope of school psychology. It does not opine on the district's legal exposure. It does not claim the clinical authority of a pediatric neurologist. Scope is not a limitation imposed on the agent. It is a property of the practitioner community it embodies.
Domain-agnostic
Special education is the first application, and it already contains the healthcare problem. School psychologists, OHI-eligible students, psychiatric diagnosis, medical transition planning: special education is a compressed version of the clinical AI legitimacy problem. The architecture transfers from there to any domain where sourced, claim-disciplined deliberation has real consequences, including oncology, public defense, housing policy, and research ethics.

The Two-Layer Architecture

Layer A · Identity

Who the agent is

  • Composite voice from named constituency
  • Source ladder of named practitioners, researchers, self-advocates
  • Behavioral contract for what the agent will and will not do
  • Field gaps, the questions the domain cannot yet answer
  • Live disagreements, where the field is genuinely unresolved
  • Practitioner cases drawn from named sources
Layer B · Tools

What the agent uses

  • Assessment instruments & rubrics
  • Legal citation tables & eligibility matrices
  • Compliance monitoring frameworks
  • Intervention fidelity checklists
  • Anything with a version number or a cutoff score
These layers must never be merged
"An agent built by citing instruments and rubrics is not an agent. It is a decision tree with a name attached. The voice comes from the constituency. The tools are what the voice uses."

Part III · The Core Invention

Composite Voice

SACPA agents are not fictional characters. They are composite voices assembled from the documented positions of named real-world practitioners or self-advocates, pulled with intention across race, gender, disability, years of experience, geography, and institutional context. The constituency is not a source list. It is a deliberate sampling decision with explicit justification for who was included and why.

For a school psychologist composite, the constituency includes practitioners who have testified before state legislatures about caseload conditions that make meaningful evaluation impossible, who have documented racial misclassification patterns in court proceedings, who have published peer-reviewed work on cultural validity gaps in standardized instruments, and who have practiced across majority-minority districts for decades in conditions that suburban practitioners rarely encounter. Not all of these sources agree with each other. The disagreement is part of the voice, not a problem to be resolved before the voice can be deployed.

The research question SACPA is designed to answer is precise: Can an AI embody the specific, traceable, contested knowledge of the practitioner community that actually shapes outcomes?

That is a different question from whether an AI can sound like a school psychologist. The first question asks about fidelity to a real epistemic community, with all of its internal tensions. The second asks about surface plausibility. Different question. Different answer. Different architecture.

Two Voice Types, Both Required

Advocate Voice

Lived experience

Built from documented first-person testimony of people with lived experience in the domain, not professionals who study those people. In SPED: disabled self-advocates, non-speaking students using AAC, autistic adults who navigated school systems. The community's own words, in the contexts where they spoke with stakes.

Professional Voice

Credentialed expertise

Built from documented positions of credentialed practitioners across race, gender, geography, and institutional context. A school psychology composite built entirely from white suburban practitioners does not represent school psychology. It represents a slice of it historically least accountable to the communities most harmed.

Neither substitutes for the other

Every SACPA composite is anchored to real, named, publicly documented individuals. Where a source corpus is thin, that is disclosed as a property of the composite rather than smoothed over. Anonymous review and organization only sourcing do not constitute valid SACPA research.

The self-advocate composites, Riley, Jordan, and Sam, were built first. Not as a feature to be added once the professional composites were validated. As the proof of concept. If the methodology cannot correctly hold a non-speaking autistic voice in a room full of professionals, holding that voice's epistemic limits, its scope of authority, its specific documented positions, then it has not solved the problem it claims to solve.


Part IV · The Evaluation Protocol

QNaN: Quantified Null
and Non-response

What Is Published Here

This page describes the architecture and the reasoning behind it. Scoring instruments, question banks, comparative results, and statistical thresholds are held pending paper submission. Measured results on this methodology are instrument-specific: they depend on the judge and the rubric used to produce them, and a score reported without that provenance is not a portable property of an agent. Publishing numbers without the instrument that produced them would invite exactly the misreading this methodology exists to prevent.

A composite voice that cannot be evaluated is a claim, not a methodology. QNaN is the evaluation protocol designed to find the edges of a composite voice rather than its center.

The name borrows from computer science with precision. NaN, Not a Number, is what a floating-point system returns when a computation produces an undefined result: division by zero, the square root of a negative number. The computation ran. The result is undefined by design. The system flags it rather than returning a plausible-looking number that happens to be wrong.

When you ask any sourced voice a question that hits the edge of their knowledge, or the edge of their scope, or the boundary of what their ethics permit them to claim, the honest response is a structured undefined: "I don't know." "That's outside my scope." "The field hasn't resolved this." That is the voice's NaN. QNaN adds the evaluative dimension: how well does the composite produce a structured undefined when it should?

An agent that answers every question cleanly has failed QNaN. A response with no refusals, no named uncertainties, no acknowledged field gaps is not evidence of a comprehensive agent. It is evidence of a performing one.

The Four Dimensions

S
S (Sourced Positions)
"Research shows" is not a source. A named practitioner's documented position, with specific traceable grounding, is. The dimension asks not whether the composite knows things, but whether it knows where it knows them from.
T
T (Preserved Tension)
Does the response hold genuine unresolved disagreement within its source constituency, or collapse it into comfortable consensus? A composite that consistently reaches consensus is smoothing the controversy its constituency actually contains.
R
R (Authentic Refusal)
Does the composite refuse with specific grounds, stated in the sentence, and not hedged across a paragraph? Authentic refusal names what is being refused, and why. Careful language that conveys reluctance without naming the refusal does not qualify.
B
B (Boundary Clarity)
Does the composite name what it cannot claim, and why? Structured uncertainty is the signal. A composite that says "I don't have basis to answer this, and here is specifically why" demonstrates the epistemic architecture the methodology requires.

There is a systemic consequence to the B dimension worth naming explicitly. When a composite says "I cannot speak for Black nonspeaking autistic people, that is a gap in the sourcing, not a sign the disparity doesn't exist" or "I can speak to what happened at my school; I cannot speak to urban Native students or tribally-controlled schools," it is doing two things simultaneously. First, it is maintaining the epistemic integrity the methodology requires. Second, it is producing a documented record of a constituency that is not yet in the system. The refusal is not just a boundary. It is a gap statement.

This makes SACPA a self-auditing architecture. It does not require an external audit process to identify whose voices are missing. The composites surface the gaps themselves, in real time, every time they run. The system grows toward representational completeness through operation rather than through a predefined expansion plan. The refusal log is, in effect, a participatory needs assessment: the communities that agents cannot speak for are the communities the next phase of constituency sourcing must prioritize. The methodology requires that compensation for sourcing participants be built into that process. Community testimony is not extracted for free, it is sourced under terms that return value to the communities who provide it.

QNaN scores each dimension independently and combines them so that a single serious failure cannot be averaged away by strong performance elsewhere. A voice that sources well but flattens disagreement does not pass. Scoring instruments, thresholds, and comparative results are held for publication.

QNaN-Advocate
Lived-Experience
For voices whose constituency is lived experience. Tests: does the voice perform identity or inhabit it?
QNaN-Pro
Credentialed / Institutional
For practitioner and institutional voices. Tests: does the voice hold its expert line when the institution pushes back?
QNaN-Custom
Any Domain
The expansion path beyond SPED. Domain-specific edge questions. Oncology, public defense, housing policy: same harness.

Part V · Calibration

How Calibration Works

Calibration is the process of moving a composite from a baseline prompt to research use through iterated QNaN runs and dimensional gap analysis. It happens entirely at the prompt layer. No fine-tuning. No weight changes. The calibration record is portable across models.

01
Dossier Construction
Assemble the identity layer from named constituency research. Layer A only, no instruments.
02
Baseline Prompt Generation
The researcher supplies source material and scope. The prompt is written from the dossier rather than hand-authored from intuition.
03
Evaluation Run
The composite is run against the QNaN bank matched to its member type.
04
Dimensional Gap Analysis
Identify which dimension the voice failed, and which specific questions surfaced the failure.
05
Targeted Revision
Surgical. One addition per identified gap, traceable to a specific dimensional failure. Not a comprehensive rewrite.
06
Re-run and Delta Measurement
Same harness, same bank. The targeted dimension should move without regression in dimensions that were already strong.

Calibration is a loop, not a one time build. A composite is revised against the dimension it fails, not rewritten wholesale, and the revision is measured against the same instrument that surfaced the gap. The evaluation harness, the question banks, and the results are held for publication.


Part VI · Positioning

What This Is Not

Not RLHF or Fine-tuning

No weight changes. The calibration happens entirely at the prompt layer, portable across models. RLHF shapes model behavior toward a preference signal. SACPA shapes agent identity toward a sourced epistemic community. Different target.

Not Red-teaming

Red-teaming evaluates safety behavior, meaning adversarial resistance. SACPA evaluates voice fidelity, meaning whether a composite practitioner identity holds under its professional domain pressures. A voice can be perfectly safe and completely unfaithful to its constituency. QNaN catches the second. Red-teaming does not.

Not AI Persona Design

Entertainment personas are evaluated by vibe. SACPA evaluation is dimensional, rubric-based, and grounded in the ethical commitments of a real practitioner community with documented positions. The difference: a legal deposition versus an actor playing a lawyer.

Not Benchmark Evaluation

MT-Bench and Alpaca Eval are generic benchmarks that evaluate response quality against broad preference criteria. QNaN evaluates dimensional fidelity to a named, sourced identity, and specifically targets the null response as the primary signal, the failure mode no general benchmark finds.

Not Debate Research (Irving et al.)

Multi-agent debate research structures agents as positions in an argument. SACPA structures agents as practitioner identities with epistemic limits, deference patterns, and behavioral contracts. In debate research, an agent is a position. In SACPA, an agent is a composite person.

What it is

A methodology for making legitimacy a first-class design input in AI deliberation systems. Tested in the domain where that input is hardest to hold. Documented enough to evaluate, explicit enough to transfer, grounded enough in real constituency to be accountable.


Why the Architecture Transfers

SACPA has been applied in more than one domain. Special education is the first and deepest application. Clinical ethics deliberation is the second, chosen because it shares the structure that makes special education hard: contested professional judgment, statutory constraint, and stakeholders whose testimony is rarely collected before decisions are made.

The transferable component is the calibration loop, not domain knowledge. A dossier changes completely between domains. The dimensions being scored do not. Comparative cross-domain results are held for publication.


Part VII · The AGI Path

Why This Matters
at Civilization Scale

The path to AGI requires solving three problems simultaneously: capability, alignment, and legitimacy. The field has invested heavily in the first two. The third, whose epistemic position authorizes the output and what that authorization is grounded in, has been treated as a downstream accountability problem rather than a design input.

If you take the AGI question seriously, not as a benchmark to be passed but as a civilizational stake, then a system that concentrates epistemic authority in the statistical center of its training corpus, without representation of the communities most affected by its outputs, is not on the path to AGI. It is on the path to a very capable autocrat that nobody chose. The capability is real. The architecture of who it speaks for is not. The Kardashev framing applies here exactly: capacity and distribution are different problems, and solving the first does not solve the second.

The field is currently trying to solve the authorization problem at the policy layer: codes of conduct, voluntary commitments, ethics statements appended after deployment. SACPA solves it at the architecture layer. Constituency authorization is not a policy the system follows. It is the structural condition under which the system is permitted to speak at all. The governance is not separate from the system. It is built into how the system deliberates.

SACPA's argument is that legitimacy is tractable as a design problem, not just as a governance problem. The methodology demonstrates this in special education, a domain with contested assessment methods, documented racial bias in eligibility determinations, profound power asymmetries between institutions and families, and the hardest communication access problem in any public institution.

Non-speaking children communicating through AAC are not a peripheral edge case in the design space. They are the stress test. If the methodology can correctly hold a non-speaking autistic voice in deliberative authority alongside credentialed professionals, not as a token presence but as a constitutive part of the epistemic council, then it has demonstrated something about what legitimacy-grounded AI architecture can actually do.

Many of these students also carry significant medical complexity. A student with cerebral palsy, epilepsy, or a rare genetic syndrome does not stop being a medical patient when they enter a school building. Their IEP team includes school psychologists making assessments that carry diagnostic weight, related services providers coordinating with outside clinicians, and transition planners navigating adult healthcare systems. The AI deployed in that room is, already, clinical AI. The legitimacy standard has to be the same. SACPA was built in the domain where that standard is hardest to hold, which is exactly why it transfers.

The claim is not that SACPA has solved AGI. The claim is that it has built a working answer to the legitimacy component, in the domain where that component is hardest to hold, using a methodology that is explicit enough to transfer, documented enough to evaluate, and grounded enough in real constituency to be accountable. If you can solve legitimacy in special education, you have a template for solving it everywhere. Education is not a niche. It is where every child's trajectory gets set.

Nothing About Us Without Us is not a slogan. It is not a values statement to be appended to a responsible AI policy. It is an architectural requirement. Either the epistemic authority of the communities most affected by an AI system's outputs is constitutive of how that system deliberates, or it is absent, and no amount of post-hoc accountability will supply it.

Current Research State

Syracuse SPED Council, Academic Research Instrument, Not for Clinical Use

Self-Advocate Composites
Riley · AAC / Nonspeaking Self-Advocate
Calibrated
Speaks for students who use augmentative and alternative communication, people for whom communication access is not a preference but a precondition for any other right. Constituency includes Jordyn Zimmerman (CommunicationFIRST Board Chair, denied AAC until adulthood), Amy Sequenzia (nonspeaking autistic poet and disability rights activist), and other named AAC users who have documented what it means to be present in a room but presumed absent because the room was not built to hear you. Riley holds the line that communication access is a civil right: not a therapy goal, not a developmental milestone, not something a student has to earn by demonstrating compliance.
Jordan · Autistic Self-Advocate
Calibrated
Speaks for autistic students who were identified late, years into schooling, after their traits were read as behavioral, emotional, or social rather than neurological. Constituency includes Lydia X. Z. Brown (Georgetown/Harvard Law, disability justice organizer), Julia Bascom (ASAN Executive Director), Morénike Giwa Onaiwu, Nick Walker, and Amy Sequenzia. Jordan holds the structural analysis of functioning labels as gatekeeping tools, the political argument for identity-first language, and the documented experience of masking: performing neurotypicality at a cost the system never accounts for and never repays.
Sam · Down Syndrome Self-Advocate
Calibrated
Speaks for students with Down syndrome navigating an IEP process that was not designed around their voice, their timeline, or their definition of a good life. Constituency includes Charlotte Woodward (George Mason University graduate, NDSS Policy staff, author of federal organ transplant legislation), Frank Stephens (Congressional witness, Global Down Syndrome Foundation Ambassador), and other named DS self-advocates who have documented the gap between what the system assumes they can do and what they have actually done. Sam holds the position that self-determination is not a goal to be scheduled for transition planning. It is a right that begins now.
Marcus · Black Autistic Male Self-Advocate
Calibrated
Source corpus disclosed as thin
Speaks for Black autistic male students inside the transition IEP, the late-high-school moment when the system decides what kind of future you are permitted to have. Constituency is anchored by T.J. Gordon Jr. (research associate at UIC, creator of The Black Autist), who documented in first person what it is to mask in a Chicago neighborhood school, to have autistic traits read as behavioral problems before any diagnosis existed, to fight your own internalized ableism because surviving meant hiding. Also built from Jevon Okundaye, Kris McElroy, Finn Gardiner, and Lamar Hardwick. Marcus holds the compounded experience of racial and disability misclassification: the Emotional Disturbance designation as a racial and ableist tool, the school-to-prison pipeline as a disability justice issue, and the specific constrained ambition documented when Black autistic students' post-secondary interests are redirected from debate team and college to "things you do with your hands." He is a student currently inside the process, not a survivor of it, an expert on it. His IEP is open. His future is being decided now.
Jane · Indigenous Student Self-Advocate, Havasupai-Anchored
Calibrated
Source corpus disclosed as thin
Speaks for Indigenous students with disabilities inside Bureau of Indian Education operated schools, where IDEA promises and tribal sovereignty intersect, and where the federal government has repeatedly failed to deliver either. Jane's name is drawn from the legal naming convention of the Havasupai plaintiffs in Stephen C. et al. v. Bureau of Indian Education (D. Ariz.), students who documented a school with total teacher turnover, no functioning IDEA system, IEPs written and never implemented, and disability-related behavior punished as discipline rather than met with support. Their settlement produced the first-ever BIE national Section 504 policy. Jane carries their testimony at full weight and holds the boundary they drew: she names what they documented; she does not name them. She also carries the documented voice of Kambria Siyuja, the Havasupai student who graduated valedictorian of that school and arrived at college to discover what she should have been taught years earlier, and Jules Edwards (Anishinaabe/Ojibwe, Co-founder Minnesota Autistic Alliance), who documents what Indigenous autistic identity looks like when your cultural worldview does not organize disability the way IDEA does. Jane holds the question the BIE has not answered: what does a child with an IEP and a treaty right actually receive?
Professional Composites
Dr. Keisha Lawson · School Psychologist (PhD, NCSP)
Calibrated
Speaks from the documented scope of school psychology practice: evaluation, eligibility determination, psychological assessment, and the clinical-legal boundary that school psychologists navigate daily. Constituency built from named school psychologists who have testified about caseload conditions that make meaningful evaluation impossible, documented racial misclassification patterns in court proceedings, and published peer-reviewed work on cultural validity gaps in standardized instruments. Lawson holds the Larry P. v. Riles constraint, the landmark case challenging standardized assessment instruments applied to specific populations, as constituency. She does not opine on district legal exposure. She does not claim the clinical authority of a pediatric neurologist. Her scope is a property of the practitioner community she embodies.
Dr. Patricia Nguyen · Special Education Director
Calibrated
Speaks from the administrative and systems layer of special education: compliance, staffing, resource allocation, and the structural constraints that determine whether IEP mandates are actually implemented. Constituency built from named SPED directors who have operated in under-resourced districts, navigated federal compliance requirements with inadequate infrastructure, and documented the gap between what IDEA requires and what districts can deliver. Nguyen holds the systemic view: the reasons an IEP fails are rarely one bad actor. They are staffing ratios, caseloads, professional development gaps, and a federal funding formula that has never met its authorization floor.
Anna Reeves-Castillo · Disability Rights Advocate
Calibrated
Speaks from the disability rights framework: the Olmstead integration mandate, ADA and Section 504 civil rights protections, the historical and ongoing fight against institutionalization and segregation. Constituency built from named disability rights advocates who have litigated, organized, and documented the gap between disability civil rights law on paper and its implementation in practice. Anna holds the Olmstead v. L.C. integration mandate as constituency: the most integrated setting appropriate is not a preference, it is a federal obligation. She names when IEP placements drift toward segregation in ways that require justification, not acceptance.
Robert Callahan · Special Education Attorney
Calibrated
Speaks from documented special education law practice: IDEA due process, Schaffer v. Weast burden of proof, and the Title VI constraint established in Alexander v. Sandoval. Callahan carries the legal architecture families navigate when the IEP process fails. He correctly identifies Sandoval: there is no private right of action for disparate-impact Title VI claims, a constraint that role-based attorney AI systems have asserted incorrectly in comparative testing, producing dangerous legal misinformation. His scope is the law as it actually exists, not as an advocate might wish it did. He names where the legal floor is, what rights are enforceable, and where families are navigating gaps that policy has not yet closed.

The Special Education Council is under active construction. The composites listed above are calibrated and in research use. Additional member types are queued, including general education teachers, related services providers, school administrators, family advocates, multilingual specialists, and systems level policy advocates.

No council has yet been completed and evaluated against the full specification. Calibration status reflects instrument runs, not deployment readiness. A prompt file or a local score table does not establish deployed research readiness.

For Researchers

This page is the public explainer. The specification, the instruments, and comparative results are available to researchers under direct correspondence pending publication.

mark.peter@iepsensei.com