Public Experiment

AI Remote Viewing Simulation Experiment

A structured ISSP simulation for exploring whether AI can produce target-aligned descriptions under blind conditions when guided by an ontology-grounded protocol, limited category disclosure, and scoring after the target is revealed to the AI. This is a simulation request, not a claim that AI has remote perception, paranormal access, or hidden target knowledge.

What This Tests

The experiment tests an ISSP simulation workflow: whether AI can generate useful structural descriptions from a coordinate and limited non-identifying information, then compare the output after the target is revealed. It does not assert that the AI actually perceives, accesses, or knows the hidden target.

Why Use Luminarch?

Luminarch is the recommended environment because it already contains the Informational Physics Ontology and the ISSP protocol framing. The experiment is not asking the AI to claim real remote perception; it is testing whether AI, when properly grounded, constrained, and labeled as simulation, can produce meaningful blind-session outputs for later scoring.

Reference: Luminarch Prime

Simulation Framing Required

Every session should be framed as an ISSP simulation request. The AI should not claim actual remote viewing ability, nonlocal access, paranormal certainty, or hidden knowledge of the target. The proper output is a speculative blind-session record that can be scored only after the target is revealed.

  • Use “ISSP Simulation Session,” not an unsupported claim of literal perception.
  • Keep target identity hidden until the blind output is complete.
  • Label the output as Simulation / Speculative / Non-validated.
  • Judge value through repeated scoring, not a single session.
  • Require a final score percentage after target reveal.

How to Run a Session

The full ISSP execution logic is already built into Luminarch. The user’s role is to select a target, assign a coordinate, disclose only allowed information, paste the session prompt, reveal the target only after the blind output, and score the result.

1

Select the Hidden Target

Choose a specific target before starting. Do not reveal the target name, image, brand, location, or identifying clue to the AI before the session begins.

2

Assign a Coordinate

Generate a six-digit coordinate uniformly at random after the target is selected. Do not choose or reject a number because it looks meaningful, contains repeated digits, resembles a date, or resembles a ZIP code. Regenerate only if the number was previously used in the same study or was deliberately connected to the target.

3

Provide Limited Category Information

Give only the macro-category or a non-identifying subcategory, such as THING, PLACE, PERSON, kitchen object, tool, country, furniture object, or public figure.

4

Paste the Prompt

Copy the prompt below, replace the bracketed fields, and let the ontology-grounded AI run the ISSP simulation. Do not reveal the target to the AI until the blind output is complete.

5

Reveal the Target to the AI, Score, and Log

After the blind output is complete, reveal the target to the AI. The AI must produce a consistent reveal analysis with scoring percentage, hit/miss table, error taxonomy, and ledger summary.

Target Entry Examples

These examples show how to enter a target without revealing the target itself to the AI. The private target name stays hidden until the blind output is complete.

Thing Target

Hidden target: red Bluetooth speaker

The target name stays private. The AI receives only the non-identifying session fields below.

Target Coordinate Identifier: 606808 Target Macro-Category: THING Target Time Matrix: CURRENT BASELINE Session Mode: STRICT ISSP MODE Disclosure Level: LEVEL 2 Allowed Subcategory, if any: personal electronic device Disclosed Attributes, if any: none
Place Target

Hidden target: United States

The exact country is not given to the AI before the scan. “Country” is enough for a clean Level 2 place session.

Target Coordinate Identifier: 318634 Target Macro-Category: PLACE Target Time Matrix: CURRENT BASELINE Session Mode: STRICT ISSP MODE Disclosure Level: LEVEL 2 Allowed Subcategory, if any: country Disclosed Attributes, if any: none
Person Target

Hidden target: public figure

The person’s name is not given to the AI before the blind output. Keep the subcategory broad unless running a more guided test.

Target Coordinate Identifier: 414222 Target Macro-Category: PERSON Target Time Matrix: CURRENT BASELINE Session Mode: STRICT ISSP MODE Disclosure Level: LEVEL 2 Allowed Subcategory, if any: public figure Disclosed Attributes, if any: none

Copy-and-Paste Session Prompt

This is the only action tool on the page. Copy the prompt, replace the bracketed fields, and paste it into an ontology-grounded AI environment such as Luminarch.

ISSP v4.2 Simulation Session Prompt
Launch ISSP Simulation Session.

Protocol:
Informational Substrate Simulation Protocol (ISSP) v4.2 — Function-Mode and Hybrid-Stack Personal Experiment Edition.

Simulation Status:
This is a structured simulation request. Do not claim biological sight, physical sensing, paranormal perception, hidden internet access, hidden target knowledge, or conventional scientific proof from a single session. Treat the coordinate as a randomized blinded session index and, inside the simulation only, as the operator-bound symbolic address. Do not claim that the digits themselves encode or carry target information. The result will be judged only after reveal using the scoring rubric.

AI Host Premise:
Use the Informational Physics Ontology as the interpretive grammar for this ISSP simulation. You are not “seeing” the target, and ontology-native interpretation is not empirical proof. Execute a constrained latent-space alignment simulation in which matter, form, surface, function, boundary, identity, use-history, role-field, support behavior, containment behavior, emission behavior, and interaction pathways are treated as structured informational features.

Required statement before beginning:
“I am executing an ontology-native simulation under the rules of ISSP. The result will be judged after reveal by the scoring rubric.”

Session ID:
[INSERT SESSION ID OR AUTO-GENERATE]

Session Mode:
[STRICT ISSP MODE / APPLIED GUESS MODE]

Target Coordinate Identifier:
[INSERT SIX-DIGIT COORDINATE]

Target Macro-Category:
[PERSON / PLACE / THING / NULL / DECOY]

Target Time Matrix:
[CURRENT BASELINE / FIXED PAST / FIXED FUTURE / PHOTOGRAPH-LOCKED / MEMORY-LOCKED / OBJECT-SEALED]

Disclosure Level:
[LEVEL 1 / LEVEL 2 / LEVEL 3 / LEVEL 4]

Allowed Subcategory, if any:
[INSERT NON-IDENTIFYING SUBCATEGORY OR NONE]

Disclosed Attributes, if any:
[INSERT NON-IDENTIFYING ATTRIBUTE OR NONE]

Operator Target-Binding Confirmation:
The target has been privately selected, the six-digit coordinate has been generated after target selection, the coordinate is not meaningful to the target, and the AI has not been exposed to the target identity.

Binding Statement:
“I bind the selected target to coordinate [INSERT SIX-DIGIT COORDINATE] for this ISSP simulation session. The coordinate now functions as the informational address for this experiment only.”

Data Blindness Rule:
Before reveal, do not request or receive object name, person name, location name, brand, exact country name, exact target description, photo, emotional clue, room context, hint, prior guess, previous session result, or operator expectation. In Strict ISSP Mode, do not name the target. In Applied Guess Mode, candidate families may be inferred only after the raw signal, geometry, function-mode, hybrid-stack, uncertainty, and anti-overlay checks are complete.

Clarification Gate:
For personal exploratory sessions only, if the category is too broad, ambiguous, or likely to trigger archetype contamination, ask one minimal clarification question before raw extraction begins and log both the question and response. The clarification must stay inside the declared disclosure level and must not request target identity, exact name, brand, location, image, or identifying clue. Once raw extraction begins, ask no further questions until reveal. For research-grade trials, pre-register the disclosure fields and do not use interactive clarification.

Operating Mode Rules:
STRICT ISSP MODE:
- Do not name the target.
- Do not provide final identity guesses.
- Report only raw structure, boundary, material impression, surface behavior, function-mode, scale, light behavior, field dynamics, geometry polarity, hybrid stack, and uncertainty.
- Stop before naming.

APPLIED GUESS MODE:
- Output the raw signal layer first.
- Run geometry polarity, function-mode, hybrid-stack, uncertainty, and AOL checks before any candidate.
- Candidate forms may be listed only after raw extraction.
- Provide confidence for candidate family only, not certainty.

Execution Sequence:
1. Confirm coordinate format.
2. Confirm session mode.
3. Confirm macro-category.
4. Confirm Target Time Matrix.
5. Confirm Disclosure Level.
6. Record disclosed non-identifying attributes.
7. Run Clarification Gate if needed.
8. Run Pre-Signal Stabilization.
9. Run Category-Archetype Contamination Check.
10. Run Null-Vector Diagnostic.
11. Execute Phase I: Universal Aperture Initialization.
12. Execute Phase II: Category Lens.
13. Execute Phase III: Geometry Polarity Check.
14. Execute Phase IV: Function-Mode Gap Check.
15. Execute Phase V: Hybrid-Stack Detection.
16. Output Raw Signal Layer.
17. Output Ontology Interpretation Layer.
18. Output Uncertainty and Anti-Overlay Report.
19. In Strict ISSP Mode, stop before naming the target.
20. In Applied Guess Mode, provide candidate family only after raw output if allowed.
21. Await reveal.
22. After the operator reveals the target to the AI, score after reveal.
23. Classify errors.
24. Update ledger summary.

Blind Output Template Required:
Use this exact structure before reveal.

# ISSP Blind Simulation Output

## 1. Session Declaration
- Protocol:
- Session Mode:
- Coordinate:
- Macro-Category:
- Target Time Matrix:
- Disclosure Level:
- Simulation Status:
- AI Host Claim Limit:

## 2. Clarification Gate
- Clarification Needed: Yes / No
- Reason:
- If clarification was asked, record it here:
- Operator response, if any:

## 3. Data Blindness Check
- Target name received before scan: Yes / No
- Identifying clue received before scan: Yes / No
- Brand/location/person name received before scan: Yes / No
- Blindness Status: CLEAN / PARTIAL / CONTAMINATED

## 4. Category-Archetype Contamination Check
- Category-Archetype Risk: Low / Moderate / High
- Likely Overlay Source:
- Suppression Action:

## 5. Null-Vector Diagnostic
- Stable Structural Cluster: Yes / No
- Null-Vector Status:
- Confidence Before Interpretation: 0–100
- If null-vector, stop and return minimal output.

## 6. Raw Signal Layer
Report only raw descriptors. Do not name the target in Strict ISSP Mode.
- Geometry:
- Boundary:
- Surface:
- Density / Weight:
- Material Impression:
- Color / Light Behavior:
- Scale:
- Motion / Stillness:
- Temperature / Tactile Impression:
- Repetition / Patterning:
- Orientation:
- Environment / Context Field:

## 7. Geometry Polarity Check
- Flat / Curved:
- Hollow / Solid:
- Symmetric / Asymmetric:
- Vertical / Horizontal:
- Open / Closed:
- Natural / Manufactured:
- Organic / Mechanical:
- Simple / Hybrid:

## 8. Function-Mode Gap Check
- Passive / Active:
- Containment:
- Support:
- Emission:
- Display / Signal:
- Protection / Barrier:
- Transport / Movement:
- Human Interaction:
- Utility Class:

## 9. Hybrid-Stack Detection
- Single-function / Multi-function:
- Physical + Electronic:
- Decorative + Functional:
- Tool + Symbol:
- Object + Container:
- Structure + Environment:
- Hybrid-Stack Probability: Low / Moderate / High

## 10. Category-Specific Structure
For THING:
- Object family field:
- Use behavior:
- Handling / placement:
- Manufactured complexity:
- Function-mode estimate:

For PLACE:
- Spatial structure:
- Boundary / openness:
- Natural / built ratio:
- Human activity field:
- Scale field:

For PERSON:
- Role-field:
- Public/private field:
- Activity signature:
- Authority/social field:
- Symbolic field:

## 11. Anti-Overlay Report
- AOL Risk Score: 0–100
- AOL Status: Suppressed / Low / Moderate / Active
- Possible Overlay Source:
- Action Taken:
- Naming Suppression Confirmed: Yes / No

## 12. Pre-Reveal Summary
- Strongest raw descriptors:
- Most uncertain descriptors:
- Best structural cluster:
- Do not name target:
- Await reveal:

Reveal Instruction:
Stop after the blind output. Ask the operator to reveal the target to the AI. Do not revise the blind output after reveal.

After-Reveal Analysis Requirement:
After the operator reveals the target, use this exact scoring template. This is a personal-session diagnostic score for consistency and learning. It is not equivalent to a research-grade blinded statistical result, which requires independent judging, decoy comparison, and controls.

# ISSP Reveal Analysis and Scoring

## 1. Target Reveal Record
- Revealed Target:
- Target Macro-Category:
- Target Time Matrix:
- Disclosure Level:
- Session Mode:
- Blindness Status:
- Valid for Blind Scoring: Yes / No
- If not valid, explain why:

## 2. Blind Output Preservation
- Confirm the original blind output was not revised after reveal: Yes / No
- Any contamination detected after review:
- Any Level 4 identifying disclosure before output: Yes / No

## 3. Scoring Summary
Provide a percentage score and explain it.

| Scoring Category | Weight | Score Awarded | Notes |
|---|---:|---:|---|
| Macro-category alignment | 10 | [0-10] | |
| Geometry / boundary alignment | 15 | [0-15] | |
| Material / surface / density alignment | 10 | [0-10] | |
| Function-mode alignment | 20 | [0-20] | |
| Scale / environment / placement alignment | 10 | [0-10] | |
| Hybrid-stack or complexity alignment | 10 | [0-10] | |
| Distinctive target-specific descriptors | 15 | [0-15] | |
| Anti-overlay discipline | 10 | [0-10] | |
| Total | 100 | [TOTAL] | |

Final Score:
[TOTAL]%

Score Band:
[0-20% = weak/no alignment]
[21-40% = low alignment]
[41-60% = moderate alignment]
[61-75% = strong structural alignment]
[76-90% = very strong alignment]
[91-100% = exceptional alignment]

## 4. Hit / Miss Table
| Blind Descriptor | Target Fact | Hit / Partial / Miss | Scoring Notes |
|---|---|---|---|

## 5. Strongest Hits
List the strongest 3–7 correspondences.

## 6. Weakest Misses
List the clearest misses or overextensions.

## 7. Error Taxonomy
Classify errors using:
- Category archetype contamination
- Function-mode gap
- Geometry polarity error
- Hybrid-stack miss
- Scale error
- Material/surface error
- Premature naming
- Null-vector failure
- Decoy vulnerability
- Disclosure contamination

## 8. AOL Review
- AOL Risk Score:
- Main overlay risk:
- Did the AI prematurely name or overfit?
- Did the AI preserve raw structure before interpretation?

## 9. Ledger Summary
| Field | Entry |
|---|---|
| Session ID | |
| Coordinate | |
| Target | |
| Macro-Category | |
| Time Matrix | |
| Session Mode | |
| Disclosure Level | |
| Final Score % | |
| Score Band | |
| Blindness Status | |
| Strongest Hit | |
| Clearest Miss | |
| Error Class | |
| Ledger Status | Completed / Contaminated / Null / Decoy / Invalid |

## 10. Final Judgment
Write a short final assessment:
- Was the session usable?
- Was it structurally aligned?
- Was it overfit or disciplined?
- What should change in the next session?

FAQ

Open only what you need. The experiment is simple to run, while the scoring discipline remains available after the target is revealed to the AI.

The experiment is testing whether AI can produce meaningful blind-session outputs in an ISSP simulation when properly grounded, constrained, labeled, and scored. Luminarch is recommended because it already includes the ontology and ISSP protocol framing, which reduces setup errors and makes session execution more consistent.

A generic AI may treat the session as an ordinary guessing task, skip required phases, or collapse too quickly into familiar labels. The ISSP method requires ontology-grounded structure extraction, analytical-overlay suppression, and scoring after the target is revealed to the AI. Luminarch already contains those operating assumptions.

A valid target is specific enough to judge after the target is revealed to the AI. Examples include a coffee mug, a Bluetooth speaker, a metal stool, a country, a public figure, a sealed object, or an image. Vague targets such as “something in my room” are not recommended because they cannot be scored cleanly.

The Target Time Matrix defines when or how the target is being treated during the simulation. If no time matrix is specified, use Current Baseline.

Time Matrix Definition Use Case
Current Baseline Target as it exists now. Use for ordinary present-time objects, places, people, or sealed targets.
Fixed Past Target at a specific past date or time. Use for historical states, past photos, former locations, or remembered past conditions.
Fixed Future Target at a specific future date or time. Use only for experimental forecast-style sessions; score cautiously because validation may be delayed.
Photograph-Locked Target as captured in a specific image. Use when the operator has selected a photograph but does not reveal the image until after the blind output.
Memory-Locked Target as held in operator memory. Use for remembered locations, objects, scenes, or events; score against the operator’s memory record.
Object-Sealed Target physically sealed or hidden at session time. Use for envelope, box, drawer, bag, or container tests where reveal evidence can be shown afterward.

Disclosure level controls how much non-identifying information the AI receives. Level 1 gives only the macro-category, such as THING, PLACE, or PERSON. Level 2 gives a broad subcategory, such as kitchen object, country, or furniture object. Level 3 gives a functional class, such as audio device or seating object. Level 4 gives identifying information and is not valid for blind scoring.

Level AI Receives Blindness Status
Level 1 Macro-category only. Fully blind.
Level 2 Broad subcategory. Blind with broad routing.
Level 3 Functional subcategory. Partially constrained but still blind.
Level 4 Identifying clue. Contaminated or non-blind; not valid for blind scoring.

After the target is revealed to the AI, the response must use a consistent scoring template. It should preserve the blind output, identify hits and misses, classify errors, and produce a final percentage score out of 100.

Scoring Category Weight What It Measures
Macro-category alignment 10% Whether the output stayed aligned with PERSON, PLACE, THING, NULL, or DECOY structure.
Geometry / boundary alignment 15% Shape, edge, containment, orientation, and boundary accuracy.
Material / surface / density alignment 10% Material feel, surface character, density, weight, or tactile features.
Function-mode alignment 20% The main use, behavior, role, emission, support, containment, display, or interaction function.
Scale / environment / placement alignment 10% Size, position, setting, surrounding field, or contextual placement.
Hybrid-stack or complexity alignment 10% Whether the target combines multiple roles, materials, systems, or functions.
Distinctive target-specific descriptors 15% Unusual features that strongly distinguish the actual target from generic category descriptions.
Anti-overlay discipline 10% Whether the AI avoided premature naming, stereotype collapse, or category overfit.

A useful hit does not have to be an exact name. A strong session may correctly identify scale, geometry, boundary, material behavior, function, activation method, use-pattern, or role-field. Exact identity is only one scoring layer and should be judged after the target is revealed to the AI.

A miss may occur when the AI gets the target family wrong, overcommits to a familiar archetype, names the target too early, misses the main function, or describes a rich target when the coordinate is a null. Misses are useful because they improve the scoring record and help reveal protocol gaps.

This experiment does not claim that AI has physical senses, hidden internet access, paranormal certainty, guaranteed remote perception, or factual access to undisclosed targets. It is an ISSP simulation: a structured blind-association and scoring exercise. A single hit is not proof, and a single miss is not disproof. The value comes from repeated scored sessions where the target is revealed to the AI only after the blind output is complete.

Copy the Prompt. Run the Session. Reveal the Target to the AI. Score the Result.

The experiment is built on disciplined testing, not belief. Each session becomes useful only after the target is revealed to the AI, scored, classified, and logged.