Frequently Asked Questions
The Reconstruction Requirement
Every existing verification system measures what can be produced. Production can be simulated. Persistence cannot. This FAQ exists because that distinction still matters.
Definition and Nature
What is the Reconstruction Requirement?
The Reconstruction Requirement is the only verification standard that still measures what verification was always designed to measure: that structural comprehension exists independently of the assistance that may have been used to produce it.
It is not a method. It is not a stricter examination. It is a boundary — the minimum condition below which what is called verification is not verification but its simulation.
The canonical definition: structural comprehension is verified if and only if it can be reconstructed from first principles, without assistance, after a minimum of ninety days of temporal separation, in a context that differs meaningfully from the original acquisition context.
What is the difference between the Reconstruction Requirement and the Reconstruction Moment?
The Reconstruction Requirement is the standard — the specification of what must be present for verification to be valid.
The Reconstruction Moment is the test — the specific epistemic event that occurs when the conditions of the standard are met and structural comprehension either reveals itself or reveals its absence.
The requirement defines what counts. The moment reveals what exists. One is a standard. The other is the instrument that administers it.
Why does this standard exist?
Because AI assistance broke the correlation that every existing verification system depends on.
Before AI assistance was ubiquitous, producing the signals of genuine structural comprehension required the cognitive work that produces structural comprehension. The examination was reliable evidence of genuine understanding because passing it required the cognitive work that genuine understanding requires. That correlation no longer holds.
AI can produce every signal of genuine structural comprehension — coherent reasoning, accurate analysis, domain-specific sophistication, appropriate uncertainty — without any structural comprehension being present. Every verification system that tests those signals is now measuring something AI assistance can produce on demand while claiming to measure something it cannot.
The Reconstruction Requirement is not a new standard. It is the rediscovery of the only one that ever mattered — formalized now because the conditions that once enforced it naturally have been removed.
Is this a stricter version of existing tests?
No. This is the point at which testing begins to measure what it has always claimed to measure.
Existing tests measure what can be produced with assistance present. The Reconstruction Requirement measures what exists when assistance is absent. These are not stricter and more lenient versions of the same measurement. They are categorically different measurements — one of which AI assistance can defeat and one of which it cannot.
The protocol is not harder than existing assessments. It is the point at which validity returns.
The Conditions
Why are there exactly three conditions?
Because each condition eliminates one specific mechanism through which borrowed explanation sustains the appearance of structural comprehension. Remove any one and that mechanism survives. The test then certifies the thing it was designed to detect.
Temporal separation eliminates memory and residual familiarity. Assistance removal eliminates access. Novel context eliminates pattern repetition within a familiar distribution. Together they create the only conditions under which structural comprehension and borrowed explanation diverge completely. Separately, none of them is sufficient.
A condition is part of this protocol only if removing it allows borrowed explanation to satisfy the test. All three conditions meet this criterion.
Why ninety days?
Because thirty days tests memory and ninety days tests structure. The distinction between these is the distinction between what can be temporarily retained and what was genuinely built.
Short-term memory, residual contextual familiarity, and pattern recall from recent exposure can sustain the appearance of structural comprehension for weeks without any genuine structure being present. Ninety days removes these confounders reliably enough to distinguish structural persistence from retention.
What has not survived ninety days cannot be claimed as structural comprehension. What has survived is the structural residue that genuine cognitive encounter produces — independent of the recency of the encounter and independent of the assistance that may have been available during it.
Time is the only adversary borrowed explanation cannot defeat.
Why must all assistance be removed?
Because the boundary between internal structural capacity and external access is precisely the boundary the protocol is testing. If any assistance remains, the test has not been administered.
Assistance during reconstruction does not make the test harder. It eliminates the test entirely. What is observed with any assistance present is the combined performance of the practitioner and the system — which proves nothing about whether the practitioner’s independent structural comprehension exists.
The question is not how well the practitioner performs with assistance. That can be observed continuously. The question is what exists when assistance ends. That can only be observed when assistance has ended.
Why must reconstruction occur in a novel context?
Because familiarity is not comprehension. Pattern repetition within a known distribution can sustain the appearance of structural comprehension without the structural model being present.
A practitioner who has borrowed explanation can reproduce that explanation within familiar territory indefinitely. The borrowing is undetectable within the original distribution. Only outside that distribution — in genuinely novel contexts that require the structural model to adapt rather than repeat — does the difference between genuine structural comprehension and borrowed explanation become visible.
Novel context is not an additional difficulty. It is the domain where the measurement becomes valid.
Can the conditions be modified for institutional convenience?
No.
An implementation that modifies any of the three conditions is not implementing the Reconstruction Requirement. It is implementing a different measurement that borrowed explanation can satisfy — which is exactly the measurement failure the protocol exists to correct.
Configurability is the mechanism by which standards fail. A configurable condition is not a protocol condition. It is a preference.
The Two Outcomes
What are the two outcomes?
Reconstruction — structural comprehension is present, rebuilds itself from first principles, adapts to the novel context. The structure returns because it exists.
The Gap — structural comprehension is absent. Fragments may be recalled. No architecture emerges. The first step does not generate the second. Nothing returns because nothing was built.
Two outcomes. No intermediate state. Both provide accurate information.
Is The Gap a failure?
No. This distinction matters and is worth addressing directly.
The Gap reveals that structural comprehension was not built — but this is not a failure of the person encountering it. Borrowed explanation and genuine structural comprehension feel identical in the moment of production. The satisfaction signal fires either way. There is no contemporaneous experience that reliably distinguishes them. The Gap is the first moment where the distinction becomes visible.
The Gap is not a verdict. It is information — the most accurate information available about what was genuinely built and what was always borrowed. It is the specific, honest starting point for genuine formation.
The failure, if there is one, belongs to the systems that certified explanation quality as evidence of structural comprehension after the correlation between them had broken.
Objections
Are we already certifying what we cannot verify?
Yes.
Every credential issued through contemporaneous performance assessment in the AI era certifies what can be produced with assistance present. None of them verify whether what was produced reflects structural comprehension that exists independently. The certification is real. What it certifies is not what it claims to certify.
If performance is correct, why does reconstruction matter?
Because correctness does not fail where it is tested. It fails where it is needed.
The situations that matter most professionally and institutionally are those where established reasoning stops working — where the novel situation falls outside the distribution, where the AI-generated output has crossed the boundary of its validity, where the correct response to yesterday’s question has become the wrong response to today’s.
Correct performance within known conditions provides no evidence that this boundary can be recognized. Reconstruction is the only verification that tests whether the structural model required to recognize it exists.
Can AI-assisted understanding still be real understanding?
Only if it persists when AI assistance is removed.
If structural comprehension disappears when assistance disappears, it was not structural comprehension. It was access. The two are indistinguishable under contemporaneous assessment and completely distinguishable under the conditions of this protocol. That is precisely why this protocol exists.
Is this practical to implement?
The question is not whether the protocol is convenient to implement. The question is whether operating on capability that has never been independently verified is acceptable.
Every organization already operates under constraints. This protocol introduces one: that claimed structural comprehension must be verifiable without the assistance that may have produced its appearance. If that verification cannot be performed, the claim cannot be made. If the claim cannot be made, the credential, certification, or capability assessment that depends on it is certifying something other than what it claims.
The practical question is not whether organizations can afford to implement this protocol. It is whether they can afford not to.
Are we already relying on people whose structural comprehension has never been verified?
Yes.
Every practitioner credentialed through contemporaneous performance assessment in the AI era has been certified on the basis of a measurement that cannot distinguish structural comprehension from AI-assisted performance. This is not a statement about any individual practitioner. It is a structural statement about what existing certification systems can and cannot detect.
The absence of verified structural comprehension does not appear during normal operations. It appears at the novelty threshold — where established reasoning fails and the structural model must exist independently. That is where the cost of unverified capability is paid.
What happens if this is not adopted?
Nothing changes — until the first situation where it matters.
The absence of verified structural comprehension accumulates silently. The practitioner who borrowed all their understanding performs identically to the practitioner with genuine structural comprehension in every situation the training distribution anticipated. The divergence appears only in the situations it did not anticipate.
Those are the situations where expertise is most consequential and most trusted. And those are the situations where the absence of genuine structural comprehension, having never been detected, produces consequences that no monitoring system predicted because no monitoring system was measuring what had been absent.
A system that relies on unverified structural comprehension is not operating on capability. It is operating on assumption. Assumptions are not failures. They are invisible until they are.
And you will not know where until it fails.
Can this be partially implemented?
No.
Removing any condition converts this protocol into a different measurement — one that borrowed explanation can satisfy. Partial implementation is not reduced rigor. It is loss of validity. An institution that claims to implement the Reconstruction Requirement while relaxing any of the three conditions is not implementing a flexible version of this standard. It is implementing a different measurement while using this standard’s name.
Institutional and Organizational Questions
Who should implement this protocol?
Any entity that makes claims about the structural comprehension of its practitioners — educational institutions, professional licensing bodies, certification systems, and organizations deploying practitioners in domains where the consequences of absent structural comprehension are consequential.
The protocol does not require all claims about capability to be verified through reconstruction. It requires that claims specifically about structural comprehension — the ability to recognize when established reasoning fails, to navigate genuinely novel situations, to identify when AI-generated outputs have become invalid — be verified through conditions that can actually detect its presence or absence.
What does implementation require?
Three things: the conditions, the audit, and the records.
The conditions must be met in full: ninety days of temporal separation, complete assistance removal, genuinely novel reconstruction context.
The audit must be performed: an independent verification that the conditions were met and that the reconstruction attempt constitutes genuine reconstruction rather than pattern repetition or retrieval.
The records must be maintained: documentation that the protocol was administered, the conditions were satisfied, and the audit was completed. These records constitute the only valid evidence that claimed structural comprehension has been verified rather than assumed.
Without audit, the protocol can be satisfied on paper without being administered in reality — which is precisely the failure mode this protocol exists to prevent.
Does this replace existing assessments?
No. It completes them.
Existing assessments provide valuable information about explanation quality, reasoning sophistication, and performance under assisted conditions. What they cannot provide — and what AI assistance has made them structurally incapable of providing — is verification that the capability persists independently of the conditions under which it was demonstrated.
The Reconstruction Requirement adds the one element that transforms certification from a record of assisted performance into a verification of structural comprehension. It does not replace what exists. It supplies what is missing.
Ownership and Standard
Who owns the Reconstruction Requirement?
No one. A standard that can be owned can be weakened. This one cannot.
The Reconstruction Requirement is an epistemic reality — a logical consequence of what genuine structural comprehension is and what AI assistance has done to the relationship between its signals and its substance. It exists independently of any institution that names, formalizes, or claims it.
ReconstructionRequirement.org holds the canonical definition as open infrastructure under CC BY-SA 4.0. Any institution may implement, adapt, and build upon it freely. No institution may claim proprietary ownership, modify the conditions while retaining the standard’s name, or enclose the methodology within proprietary systems that restrict access to the standard.
The ability to verify whether structural comprehension exists cannot become intellectual property.
Can understanding be assumed?
No. It can only be reconstructed.
Where reconstruction is absent, verification does not exist. Only its appearance does.
What is the single sentence that captures the Reconstruction Requirement?
If it cannot be reconstructed without assistance, it was never understood.
That sentence is not a standard being proposed. It is a fact about structural comprehension that was always true and that AI assistance has made impossible to ignore.
ReconstructionRequirement.org — CC BY-SA 4.0 — 2026
ReconstructionMoment.org — The test through which the Reconstruction Requirement is administered
PersistoErgoIntellexi.org — The protocol that formalizes the Reconstruction Requirement
TempusProbatVeritatem.org — The foundational principle: time proves truth