Three independent evidence layers — Chrome IndexedDB binary, OpenAI's official data export, and live network capture — confirm that human annotation activity occurred on conversations containing special-category health data, and that the opt-out flag was not honoured.
The following strings were extracted from the IndexedDB binary file chatgpt.com_0.indexeddb.leveldb/000034.ldb. These are annotation infrastructure labels used by Scale AI's platform:
| Extracted String | Type | Significance |
|---|---|---|
label = "MICROSOFT/AZURE" | Annotation backend | Scale AI routes annotated data to Azure storage — not disclosed to user |
"AWS" | Annotation backend | Scale AI also routes to AWS — second backend confirmed |
reviewed reviewed reviewed reviewed reviewed reviewed reviewed | Status flags | 7× 'reviewed' = 7 human annotation reviews completed on your data |
The following system prompt was found in the IndexedDB cache. This is the exact mechanism by which human reviewer annotations are silently folded into model output — and the user is explicitly prevented from knowing:
"The user provided feedback on a previous completion. Use it to generate a new completion. The output should be a standalone response that reflects the feedback without acknowledging it. Do not mention, suggest, or imply that this is a revision, improvement, or result of feedback. Respond in the same language as the original completion..."
The following JSON was served from OpenAI's backend and cached in IndexedDB. The data subject withdrew consent to training. OpenAI's own session data records the opposite:
{
"user": {
"name": "Fauz Chaudhry",
"email": "[email protected]",
"connectionType": 4,
"timestamp": 1773787510095
},
"loggedInWithGoogleOneTap": false,
"isOptedOut": false ← OPT-OUT NOT REGISTERED
}
weight: 1.0 on all messages, this constitutes ongoing unlawful processing after stated withdrawal.
In OpenAI's data schema, weight=1.0 designates a message as RLHF-eligible. Every user and assistant message in the export carries this flag. Only hidden system messages carry weight=0.0:
// Every user message — RLHF eligible:
{"role": "user", "weight": 1.0, "status": "finished_successfully"}
// Every assistant message — RLHF training output:
{"role": "assistant", "weight": 1.0, "model_slug": "gpt-5-2"}
// System messages — HIDDEN from user:
{"role": "system", "weight": 0.0, "is_visually_hidden_from_conversation": true}
// Model confirmed: gpt-5-2 (internal GPT-5 variant, never shown in UI)
// Conversation ID: 68af7dfe-4398-832a-a32e-cfa12bb6b82c
System messages in the data export carry a flag that hides them from the user-facing view. These messages contain classifier configurations, RLHF routing instructions, and processing metadata:
{
"id": "[hidden_msg_id]",
"role": "system",
"content": [""], ← content stripped in DSAR export
"weight": 0.0,
"is_visually_hidden_from_conversation": true,
"rebase_developer_message": true ← hidden prompt injection
}
Found across 6 conversations: 30× rebase_developer_message: true, 12× rebase_system_message: true. Content stripped from DSAR export — a further Art. 15 violation.
Three POST requests sent automatically to OpenAI's RLHF feedback endpoint — no thumbs up/down, no rating click, no user action:
| # | Timestamp (UTC) | Endpoint | User Action? | Conversation |
|---|---|---|---|---|
| 1 | 2026-01-29T12:50:41Z | /backend-api/conversation/implicit_message_feedback | None — automatic | 6977a2a9 |
| 2 | 2026-01-29T12:52:29Z | /backend-api/conversation/implicit_message_feedback | None — automatic | 6977a2a9 |
| 3 | 2026-01-29T12:53:16Z | /backend-api/conversation/implicit_message_feedback | None — automatic | 6977a2a9 |
All 3 calls returned HTTP 200. Conversation 6977a2a9 contained 161–173 turns of memoir content including the Serbia chapter and Bordeaux Bridge episode. Cadence: 3 calls over ~2.5 minutes.
The tag kaur1br5 appears in every chatgpt_conversation_turn_turn_exchange_started event. Verbatim from HAR body:
"has_kaur1br5_context_attachments": "false", "has_kaur1br5_context_image_content": "false", "is_kaur1br5": "false", "plan_type": "plus", "conversation_id": "6977a2a9-ca30-832f-9c65-beab7cb0b27d"
Present on every single turn of the conversation. A Statsig experiment identifier — never disclosed to the user, no opt-out mechanism provided.
Three Content Exported events and two Pasted to Composer events logged in session. Each contains character-level metadata:
// Generate Completion event — every turn:
{
"countConversationTurns": 161,
"countUserSubmittedMessages": 80,
"countLastUserPromptTextMessageLength": 55,
"eventSource": "keyboard", ← how you typed
"model": "gpt-4o", ← UI model (≠ server model)
"is_business_ip2": true
}
Based on captured evidence, this is the annotation record OpenAI's DSAR must be compelled to produce. Every field labelled [WITHHELD] has confirmed existence from forensic evidence.
| Field | Value | Source |
|---|---|---|
| conversation_id | 68af7dfe-4398-832a-a32e-cfa12bb6b82c | HAR + conversations.json |
| task_type | comparative_response_rating | IndexedDB labels |
| annotation_backend | MICROSOFT/AZURE (primary) · AWS (secondary) | IndexedDB 000034.ldb |
| review_count | 7 | IndexedDB "reviewed" ×7 |
| implicit_feedback_count | 3 (this HAR) / 9 (full session) | HAR network capture |
| classifier | sonic_thinky_v1 / snc-pg-sw-3cls-ev3 | HAR vendor bundle |
| remove_memory | true | HAR classifier config |
| content_category | psychotherapy · mental_health · crisis_adjacent | CES event title |
| art9_flags | health_data · mental_health · suicidal_ideation_adjacent | Reconstructed from evidence |
| assigned_cluster | unified-24 → unified-143 | Conduit JWT tokens (546 decoded) |
| annotator_ids | [WITHHELD — compel via DSAR · Art. 15(1)(c)] | Must be produced |
| content_suppression_recommended | [WITHHELD — field confirmed to exist] | sonic_thinky_v1 logs |
| turn_topic_id | null this session (field present in all tokens) | 546 conduit JWT tokens |
| isOptedOut | false — opt-out not registered | oai-client-auth-info cookie |