SAMPLE · DEMONSTRATION ONLY
SAMPLE CLIENT OUTPUT — NOT AN ACTUAL COMPANY FINDING

Evidence Examination

A demonstration of the client-facing 24-point evidence report after analyst review and release. This sample uses a fictional company and representative findings solely to show format, terminology and investor-facing presentation.

Subject
NovaFoundry AI, Inc. (fictional)
Public anchor
example.invalid
Examination date
August 8, 2026
Review status
24 / 24 signed

Executive Summary

C 12R 4Q 3ND 5

This demonstration distinguishes a conforming observation, a documented risk, a substantiation gap and a point on which no issue was discovered in named reviewed sources. ND is not a claim that no risk exists outside the reviewed source set.

A — Source chain-of-custody

A1

Primary-source traceability

C · Conforms

Published technical throughput claim traces to a dated benchmark document and the referenced test record.

A2

Entity verifiability

C · Conforms

Company and named executive identities were matched to the reviewed public record.

A3

Measurable basis

Q · Substantiation gap

The site states a 42% efficiency gain but does not publish the baseline, sample, or measurement window.

A4

Reference-link integrity

R · Documented risk

One product-page citation resolves to an outdated version of the referenced technical paper.

B — Citation integrity

B1

Source–claim fidelity

C · Conforms

Reviewed source language supports the public claim at the stated strength.

B2

Internal consistency

R · Documented risk

A deployment-status statement differs between the homepage and a product brief.

B3

Headline–body match

C · Conforms

Headline language remains within the qualification stated in the body copy.

B4

Quote fidelity

ND · No issue discovered in reviewed sources

No attributed quotation requiring independent fidelity testing was discovered in the reviewed source set.

An investor would ask:
Has AI been used to transcribe, summarize, edit or generate any quotation attributed to a person?

C — Confidence-band discipline

C1

Self-report labeling

Q · Substantiation gap

Two performance statements appear company-generated but are not labeled as internal results.

C2

Hedging present

C · Conforms

Forecast and inference language is appropriately qualified.

C3

Achieved vs. projected

R · Documented risk

A roadmap result is visually close to achieved metrics and should be separated more clearly.

C4

Superlative basis

Q · Substantiation gap

The phrase “industry-leading” appears without a defined comparison set or independent basis.

D — Regulatory / compliance term accuracy

D1

Currency

C · Conforms

Reviewed regulatory references were current as of the stated examination date.

D2

Correct application

ND · No issue discovered in reviewed sources

No public representation was discovered that would support a conclusion that a specific high-risk AI classification is engaged.

An investor would ask:
Could AI be used in an internal process that changes which regulations apply to the entity or its activities?
D3

Certification accuracy

C · Conforms

Certification language is limited to the documented status and does not state completion where none is evidenced.

D4

Term precision

C · Conforms

Automation, agentic and AI terminology are differentiated with sufficient precision in the reviewed pages.

E — Brand / competitor exposure

E1

Competitor exposure

ND · No issue discovered in reviewed sources

No named competitor comparison requiring substantiation was discovered in the reviewed pages.

An investor would ask:
Has AI been used to research, compare or describe competitors in materials not visible in the reviewed public record?
E2

Disparagement / liability

C · Conforms

No unsupported negative statement about a named competitor was discovered.

E3

Third-party outcome sourcing

ND · No issue discovered in reviewed sources

No public customer outcome or testimonial was discovered that required third-party outcome verification.

An investor would ask:
Has AI created, edited or summarized customer outcomes, testimonials or case studies that are not visible in the reviewed source set?
E4

Brand / domain consistency

R · Documented risk

One legacy product domain remains active and could create identity confusion for diligence readers.

F — Attribution hygiene

F1

Name / title accuracy

C · Conforms

Named executive titles were consistent across the reviewed public pages.

F2

Speaker identifiability

C · Conforms

Synthetic media is labeled as such where used in the reviewed public presentation.

F3

No fabricated attribution

ND · No issue discovered in reviewed sources

No attributed endorsement or testimonial was discovered in the reviewed source set.

An investor would ask:
Has AI been used to draft endorsements, testimonials, quotations or statements attributed to another person?
F4

Institutional accuracy

C · Conforms

Institutional participation is described as participation and not as endorsement or approval.

Sources & custody

  1. Fictional subject-site capture and source manifest — demonstration placeholder.
  2. NIST AI 600-1 — Generative Artificial Intelligence Profile.
  3. FTC advertising substantiation and impersonation authorities — representative source categories.
  4. SEC public AI-claim enforcement materials — representative source category.
  5. Applicable jurisdictional AI-governance authority records — selected at examination time.

Signature state: All 24 points shown as analyst-reviewed for demonstration. A production report would carry the actual signed timestamp and signature-block hash emitted by the governed workflow.