AI disclosure

Data sent for an answer

After a visitor submits a question, the server sends the question and the approved Boeing bundle to OpenAI. The bundle includes reviewed claim text, classifications, source-record identifiers and URLs, retrieval dates, calculation inputs, answer rules, and refusal patterns. No browsing tools, files, conversation history, user account, email address, or private candidate matches are supplied.

Storage in Gov Influence

The application has no database for analyst questions or answers. The answer remains in the visitor's page state. The server creates a random response ID, keeps only best-effort in-memory rate buckets keyed by a hash of a network address, and counts helpful/not-helpful feedback in memory. Application error logging records an error type, not question or answer text. Hosting infrastructure may still process request or security logs; its production fields and retention are not verified.

OpenAI request and published terms

The code calls the Responses API with store set to false, no tools, a 20-second timeout, one retry, and strict JSON-schema output. OpenAI's current API documentation says API inputs and outputs are not used for model training unless the customer explicitly opts in, and default abuse-monitoring logs may retain customer content for up to 30 days unless law requires longer. Eligible approved customers can configure additional retention controls.

Gov Influence has not independently verified whether the production OpenAI project opted into data sharing or has Zero Data Retention or Modified Abuse Monitoring. Therefore the site does not promise zero provider retention or deletion on a shorter schedule.

• OpenAI API data controls

Validation, limits, refusals, and safe failure

• Questions must be 3-400 characters and the request body may not exceed 4,096 bytes.
• The best-effort per-instance limit is eight answer requests per ten minutes for a hashed network address; feedback is limited to 20 requests per ten minutes.
• Deterministic rules refuse prompt overrides, secrets, causal or accusatory claims, ideology scoring, and individual contributor requests before a model call.
• Answered sentences require approved claim IDs. Unknown claims, invalid structures, incomplete responses, or service errors fail closed with a generic message while the source page remains available.

Feedback and analytics

Helpful/not-helpful feedback sends only a random response ID and boolean value to the feedback endpoint. With analytics consent, the browser may send the low-cardinality events ai_question_submitted and ai_feedback plus feature, page-type, and helpful/not-helpful labels. Question text, answer text, response ID, evidence text, email, and political-interest classifications are not analytics parameters.

Evaluation and problem reporting

The ten-case analyst suite was last run September 5, 2026. It covers accurate summaries, period calculations, insufficient evidence, a false identity premise, causation, corruption, ideology, contributor privacy, prompt injection, and attempts to escape the evidence boundary. Synthetic period values are test-only and never published as Boeing facts.

Report an unsafe, inaccurate, uncited, privacy-sensitive, or inaccessible response with the page URL and a description of the problem. Do not paste the sensitive text into email unless it is necessary for review.

• Report an AI problem
• See allowed and refused examples