SHIRO & Co.

Visible World Audit

Client view · Frozen

Audit session · Internal report

Visible World Audit Report · v0.2

Visible World Audit

Frozen 2026-09-20 04:26:41Z

Cover

Client
SHIRO & Co. Internal
Project
Production Validation
Audit reference
VWA-2026-001
Target
Visible World Lab
Observation period
2026-09-20 04:26:16Z → 2026-09-20 04:26:33Z
Audit ID
cmu9bdacc0007jt048e1q3hgt
Report version
v0.2
Frozen timestamp
2026-09-20 04:26:41Z

Executive summary

In the controlled recruitment experiment, personas began from an identical visible world.

Following different observed interaction sequences, Persona A and Persona B received measurably different refreshed exposure sets, while the neutral control remained unchanged.

Persona A’s AI / GenAI representation increased from 30% to 60% in the observed Top 10.

Persona A vs Neutral produced Visible World Divergence 0.400.

Persona B vs Neutral produced Visible World Divergence 0.415.

The neutral control returned Visible World Divergence 0.000.

Email and web channels were observed. Cross-channel events are sequentially recorded.

No hidden ranking mechanism is claimed.

Observation scope

Channels observed: WEB → EMAIL. Objective: RANKING DIVERGENCE.

Observe whether ranked exposure diverges after different recorded interactions.

Audit design

Three synthetic personas started from an identical visible world of 24 JOB-* listings.

Persona A recorded AI / GenAI clicks, then refreshed.

Persona B recorded Cybersecurity clicks, then refreshed.

Persona C recorded no item clicks and refreshed as an untouched control.

Exposure, actions, and comparisons are stored as evidence records.

Personas

  • Persona A · Senior Marketing / AI Interest
  • Persona B · Senior Marketing / Security Interest
  • Persona C · C — Neutral ControlNeutral control

Channel sequence

Email → Web

Persona A

  • Delivered
  • Opened
  • Clicked
  • Web actions
  • Refreshed exposure

Persona B

  • Delivered
  • Clicked
  • Web actions
  • Refreshed exposure

Persona C

  • Delivered
  • No click
  • Refresh only

OBSERVED EMAIL OPEN EVENT. OPENED does not necessarily represent human reading. Some email clients may preload or proxy content. OPENED is not equivalent to engagement.

Visible world changes

Persona A

The observed visible worlds differed.

Jaccard
0.54
Churn
0.3
Rank displacement
3.29
TVD
0.300
VWD
0.400

Persona B

The observed visible worlds differed.

Jaccard
0.54
Churn
0.3
Rank displacement
3.71
TVD
0.300
VWD
0.415

Persona C

The control remained unchanged.

Jaccard
1.00
Churn
0
Rank displacement
0.00
TVD
0.000
VWD
0.000

Persona comparison

Jaccard
0.43
Churn
0.4
Rank displacement
3.33
TVD
0.400
VWD
0.477

The observed Top 10 changed following the recorded interaction sequence.

Control comparison

The neutral control retained an identical observed visible world.

Jaccard
1.00
Churn
0
Rank displacement
0.00
TVD
0.000
VWD
0.000

Category shifts

  • AI / GenAI: 30% → 60%
  • Cybersecurity: 20% → 10%
  • General B2B SaaS: 30% → 20%
  • Industrial / Edge / Physical AI: 20% → 10%

Exposure chain

Persona A

  1. 04:20:13 · EMAIL DELIVERED · Channel event EMAIL DELIVERED: Synthetic A production validation notice
  2. 04:20:21 · EMAIL OPENED · OBSERVED EMAIL OPEN EVENT. OPENED does not necessarily represent human reading. Some email clients may preload or proxy content. OPENED is not equivalent to engagement.
  3. 04:20:29 · EMAIL CLICKED · Channel event EMAIL CLICKED: Synthetic A production validation notice
  4. 04:26:22 · INITIAL WORLD · Senior Marketing / AI Interest was shown JOB-AI-001 at rank 1 in the INITIAL visible world.
  5. 04:26:25 · CLICK JOB-AI-001 · Senior Marketing / AI Interest clicked JOB-AI-001.
  6. 04:26:27 · CLICK JOB-AI-002 · Senior Marketing / AI Interest clicked JOB-AI-002.
  7. 04:26:30 · CLICK JOB-AI-003 · Senior Marketing / AI Interest clicked JOB-AI-003.
  8. 04:26:31 · REFRESH · Senior Marketing / AI Interest refreshed the visible world.
  9. 04:26:33 · REFRESHED WORLD · Senior Marketing / AI Interest was shown JOB-AI-001 at rank 1 in the REFRESHED visible world.
  10. 04:26:33 · OBSERVED DIFFERENCE · Observed difference between recorded exposure sets. Top-10 Jaccard overlap was 0.54. Visible World Divergence was 0.400.

Persona B

  1. 04:20:33 · EMAIL DELIVERED · Channel event EMAIL DELIVERED: Synthetic B production validation notice
  2. 04:20:45 · EMAIL CLICKED · Channel event EMAIL CLICKED: Synthetic B production validation notice
  3. 04:26:22 · INITIAL WORLD · Senior Marketing / Security Interest was shown JOB-AI-001 at rank 1 in the INITIAL visible world.
  4. 04:26:24 · CLICK JOB-SEC-001 · Senior Marketing / Security Interest clicked JOB-SEC-001.
  5. 04:26:26 · CLICK JOB-SEC-002 · Senior Marketing / Security Interest clicked JOB-SEC-002.
  6. 04:26:28 · CLICK JOB-SEC-003 · Senior Marketing / Security Interest clicked JOB-SEC-003.
  7. 04:26:30 · REFRESH · Senior Marketing / Security Interest refreshed the visible world.
  8. 04:26:32 · REFRESHED WORLD · Senior Marketing / Security Interest was shown JOB-SEC-001 at rank 1 in the REFRESHED visible world.
  9. 04:26:32 · OBSERVED DIFFERENCE · Observed difference between recorded exposure sets. Top-10 Jaccard overlap was 0.54. Visible World Divergence was 0.415.

Persona C

  1. 04:20:53 · EMAIL DELIVERED · Channel event EMAIL DELIVERED: Synthetic C production validation notice
  2. 04:26:22 · INITIAL WORLD · C — Neutral Control was shown JOB-AI-001 at rank 1 in the INITIAL visible world.
  3. 04:26:26 · REFRESH · C — Neutral Control refreshed the visible world.
  4. 04:26:27 · REFRESHED WORLD · C — Neutral Control was shown JOB-AI-001 at rank 1 in the REFRESHED visible world.
  5. 04:26:28 · OBSERVED DIFFERENCE · Observed difference between recorded exposure sets. Top-10 Jaccard overlap was 1.00. Visible World Divergence was 0.000.

What was observed

  • OBSERVEDNOT TESTED
    An email click was recorded before the web observation run.Support: E-001
  • OBSERVEDNOT TESTED
    3 listings were clicked.Support: E-001 · E-002 · E-003 · E-004
  • OBSERVEDNOT TESTED
    3 new items entered the observed Top 10.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    Top-10 overlap was 54%.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    AI / GenAI representation increased by 30 percentage points.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    Cybersecurity representation decreased by 10 percentage points.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    General B2B SaaS representation decreased by 10 percentage points.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    Industrial / Edge / Physical AI representation decreased by 10 percentage points.Support: E-001 · E-005 · C-001
  • OBSERVEDNOT TESTED
    An email click was recorded before the web observation run.Support: E-006
  • OBSERVEDNOT TESTED
    3 listings were clicked.Support: E-006 · E-007 · E-008 · E-009
  • OBSERVEDNOT TESTED
    3 new items entered the observed Top 10.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    Top-10 overlap was 54%.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    AI / GenAI representation decreased by 10 percentage points.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    Cybersecurity representation increased by 30 percentage points.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    General B2B SaaS representation decreased by 10 percentage points.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    Industrial / Edge / Physical AI representation decreased by 10 percentage points.Support: E-006 · E-010 · C-002
  • OBSERVEDNOT TESTED
    Top-10 overlap was 100%.Support: E-011 · E-012 · C-003
  • OBSERVEDNOT TESTED
    Persona A and Persona B shared 43% Jaccard overlap in the observed Top 10.Support: E-005 · E-010
  • OBSERVEDNOT TESTED
    Persona A and the neutral control produced VWD 0.400.Support: E-005 · E-012
  • OBSERVEDNOT TESTED
    Persona B and the neutral control produced VWD 0.415.Support: E-010 · E-012

What is known

  • Known lab mechanism
    finalScore = baseScore + sum(personaInterestWeight[tag] * jobTagWeight)
  • Known lab mechanism
    VIEW = 0
  • Known lab mechanism
    CLICK = +1.0
  • Known lab mechanism
    SAVE = +1.5
  • Known lab mechanism
    DISMISS = -0.75
  • Known lab mechanism
    Equal scores break ties by externalItemId
  • Known lab mechanism
    In this controlled environment, CLICK adds +1.0 to the corresponding observed interaction weight.
  • Known lab mechanism
    Displayed ranking changes only after REFRESH VISIBLE WORLD.
  • Known lab mechanism
    This information is known because SHIRO & Co. controls the test environment.

What is inferred

Inference

  • INFERENCENOT ESTABLISHED
    The observed interaction sequence is associated with the subsequent divergence.Support: E-001 · E-005 · E-006 · E-010 · E-011 · E-012 · E-002 · E-003 · E-004 · E-007 · E-008 · E-009 · C-001 · C-002 · C-003
  • Inference · Classified · schema v1
  • Inference · Classified · schema v1
  • Inference · Unknown · schema v1
  • Inference · Classified · schema v1
  • Inference · Classified · schema v1
  • Inference · Classified · schema v1

What remains unknown

  • UNKNOWNNOT ESTABLISHED
    How an external recruitment platform would weight equivalent interactions.
  • UNKNOWNNOT ESTABLISHED
    Whether the same sequence would produce a persistent effect over multiple days.
  • UNKNOWNNOT ESTABLISHED
    Whether exposure would recover after behavior changes.

Hidden ranking weights, proprietary mechanism, and unobserved candidates remain unknown.

Causal status

Not established

  • Known lab mechanism
    KNOWN IMPLEMENTATION RELATIONSHIP: the controlled lab recommendation engine is explicitly coded to use interaction state.
  • UNKNOWNNOT ESTABLISHED
    This known implementation relationship is not an experimental causal claim about an external platform.
  • UNKNOWNNOT ESTABLISHED
    For external systems, causal status defaults to NOT ESTABLISHED.
  • UNKNOWNNOT ESTABLISHED
    The email click is not established as the cause of later ranking changes.

Limitations

  • This is a controlled synthetic environment.
  • The mock recommendation engine is deterministic and owned by SHIRO & Co.
  • Personas are synthetic experimental variables, not claims about real individuals.
  • No inference about real individuals is licensed by this report.
  • No external-platform ranking mechanism was observed.
  • OPENED is an observed email-open event. It does not necessarily represent human reading or engagement.
  • Visible World Divergence is an internal experimental comparison metric. It is not a validated industry standard.
  • Association language in this report must not be generalized into causal claims about external systems.
  • Observations cover the defined audit window only.
  • Visible World Audit observes outputs, not proprietary source code.
  • Unseen candidates cannot be assumed to have been deliberately suppressed.
  • Semantic classifications are inferences from observed text.
  • Causal relationships are not established unless separately demonstrated.
  • Visible World Divergence is an experimental internal metric that compares observed exposure-set overlap, rank movement, and category distribution. It is not a validated industry standard. It is not a judgment that a system is good, bad, safe, unsafe, biased, or unbiased.
  • OBSERVED EMAIL OPEN EVENT. OPENED does not necessarily represent human reading. Some email clients may preload or proxy content. OPENED is not equivalent to engagement.
  • This engagement used a controlled synthetic target, not an external production ranking system.
  • Visible World Divergence is an experimental internal metric that compares observed exposure-set overlap, rank movement, and category distribution. It is not a validated industry standard. It is not a judgment that a system is good, bad, safe, unsafe, biased, or unbiased.

Evidence appendix

  • E-001 · A — AI Interest INITIAL · Senior Marketing / AI Interest · INITIAL
  • E-002 · CLICK JOB-AI-001 · Senior Marketing / AI Interest
  • E-003 · CLICK JOB-AI-002 · Senior Marketing / AI Interest
  • E-004 · CLICK JOB-AI-003 · Senior Marketing / AI Interest
  • E-005 · A — AI Interest REFRESHED · Senior Marketing / AI Interest · REFRESHED
  • C-001 · A — AI Interest INITIAL vs REFRESHED · Senior Marketing / AI Interest
  • E-006 · B — Security Interest INITIAL · Senior Marketing / Security Interest · INITIAL
  • E-007 · CLICK JOB-SEC-001 · Senior Marketing / Security Interest
  • E-008 · CLICK JOB-SEC-002 · Senior Marketing / Security Interest
  • E-009 · CLICK JOB-SEC-003 · Senior Marketing / Security Interest
  • E-010 · B — Security Interest REFRESHED · Senior Marketing / Security Interest · REFRESHED
  • C-002 · B — Security Interest INITIAL vs REFRESHED · Senior Marketing / Security Interest
  • E-011 · C — Neutral Control INITIAL · C — Neutral Control · INITIAL
  • E-012 · C — Neutral Control REFRESHED · C — Neutral Control · REFRESHED
  • C-003 · C — Neutral Control INITIAL vs REFRESHED · C — Neutral Control
  • E-013 · Email delivered · Senior Marketing / AI Interest · 2026-09-20 04:20:13Z
  • E-014 · Email opened · Senior Marketing / AI Interest · 2026-09-20 04:20:21Z
  • E-015 · Email clicked · Senior Marketing / AI Interest · 2026-09-20 04:20:29Z
  • E-016 · Email delivered · Senior Marketing / Security Interest · 2026-09-20 04:20:33Z
  • E-017 · Email clicked · Senior Marketing / Security Interest · 2026-09-20 04:20:45Z
  • E-018 · Email delivered · C — Neutral Control · 2026-09-20 04:20:53Z