Validate evidence
Confirm critical facts against current system evidence, authoritative IBM or vendor documentation and applicable local standards.
Public GPT Privacy & Responsible Use
Privacy & Responsible Use for z/OS Agent Elephant
Last updated: September 6, 2026. This notice explains how Tambition treats information associated with the public z/OS Agent Elephant GPT, how ChatGPT platform processing is separate from Tambition, and the operational boundaries users should follow when working with enterprise-system information.
Key point: do not submit credentials, secrets, personal data, customer-confidential material, proprietary source code, or unmasked production evidence. Use sanitized, minimum-necessary technical information.
Scope & Roles
Tambition and OpenAI have separate roles.
This page applies to Tambition Technologies Group Inc.’s public z/OS Agent Elephant GPT and the related Tambition website pages. The GPT runs inside ChatGPT. OpenAI operates the ChatGPT platform and controls platform-level account, model, retention, training, security and data-control behavior. Tambition controls the GPT’s purpose, instructions, knowledge materials and responsible-use boundaries.
This notice supplements, but does not replace, OpenAI’s applicable terms and privacy policies.
Official OpenAI references
Conversation Access
The GPT builder cannot view individual GPT conversations through the builder interface.
OpenAI states that GPT builders cannot view individual conversations users have with their GPTs. Tambition therefore does not receive a raw transcript merely because a user chats with z/OS Agent Elephant. This does not cover information a user separately sends to Tambition by email, shares through another channel, or intentionally transmits through a future external integration.
ChatGPT Platform Processing
Platform data use depends on the user’s ChatGPT plan and settings.
For consumer ChatGPT plans, conversations may be used to improve OpenAI models depending on the user’s Data Controls choices. OpenAI states that Business, Enterprise and Edu data is not used for model training by default. Users should review their current ChatGPT Data Controls before entering sensitive business information.
Temporary Chat and other platform controls are OpenAI features and remain governed by OpenAI’s current product behavior and policies.
User controls to review
- Model-improvement / training preference
- Chat history and deletion controls
- Temporary Chat availability and behavior
- Workspace policies and administrator controls
- Any app or action approval shown in ChatGPT
Information Users Should Not Submit
Sanitize inputs and disclose only the minimum necessary.
Users are responsible for confirming that they are authorized to disclose information they submit. For enterprise support, use reduced, masked and generalized evidence whenever practical.
- Passwords, credentials, access tokens, API keys or private keys
- Personal data or regulated personal information
- Customer-confidential or contract-restricted information
- Unmasked logs, dumps, traces, screenshots or console captures
- Production host names, IP addresses, account IDs or other sensitive identifiers
- Proprietary source code, licensed software content or restricted datasets
- Security-sensitive configuration or vulnerability information not authorized for disclosure
- Any material you do not have authority to share with ChatGPT
Tambition Website Data Handling
The current Tambition website is static and does not include a data-entry form or analytics integration.
The website does not intentionally collect GPT conversation content, uploaded files or form submissions. Ordinary infrastructure logs, browser behavior, hosting records or future analytics may be handled by the hosting or infrastructure provider and must be evaluated separately if the deployment configuration changes.
Knowledge & Experience Learning Database
The learning database is intended to hold curated technical knowledge and generalized experience, not raw public-GPT chats.
Tambition’s governance requires reusable material to be sanitized, generalized and provenance-aware. Raw customer case data, credentials, personal data, proprietary code and unmasked production evidence should not be placed into the Knowledge & Experience Learning Database. Material is not treated as reusable merely because it appeared in a GPT conversation.
SanitizeRemove secrets, identities and unnecessary production detail.
GeneralizeExtract reusable technical patterns rather than customer-specific facts.
ReviewCheck provenance, rights, accuracy and operational risk.
GovernReuse only within Tambition’s approved knowledge-governance boundary.
Apps, Actions & Future Integrations
External integrations create a separate data-sharing boundary.
If a GPT uses an app or external API, OpenAI states that relevant parts of a user’s input may be sent to that third party. Users should review any ChatGPT approval prompt and the third party’s privacy terms before proceeding.
This website does not represent z/OS Agent Elephant as having autonomous production-system authority. Any future integration with enterprise systems, repositories, monitoring, ticketing or other services requires explicit authorization, access controls and updated privacy disclosure.
Before approving an external action
- Confirm the destination system and provider
- Confirm exactly what data will be transmitted
- Remove unnecessary secrets and identifiers
- Confirm your organization permits the transfer
- Review the receiving service’s retention and privacy policy
Responsible Technical Use
z/OS Agent Elephant provides AI-assisted technical reasoning, not autonomous production authority.
Human review
Qualified personnel remain responsible for technical decisions, change approval, implementation, rollback and production verification.
Controlled execution
Test proposed changes in an appropriate controlled environment and follow the organization’s change-management, security and recovery procedures.
Accuracy, Security & Retention
AI output can be incomplete or wrong; no online service eliminates risk.
Do not rely on z/OS Agent Elephant as the sole source for a production-critical decision. Tambition does not promise that AI-generated output is complete, current, error-free or suitable for a specific environment.
OpenAI controls retention of ChatGPT platform data under its current policies and account/workspace settings. Tambition-controlled retention applies only to information Tambition separately receives or deliberately creates outside the ordinary GPT-builder conversation view.
Practical safeguards
- Minimize data before submission
- Mask identifiers and remove secrets
- Keep authoritative evidence outside the chat when required by policy
- Do not use GPT output as an unreviewed change command
- Retain approved operational records in the organization’s authorized systems of record
Questions, Corrections & Policy Changes
Contact Tambition for the Tambition-controlled part of this service.
For questions about this notice, Tambition-controlled website content, or correction of Tambition-published material, contact tambition@tambition.com. Questions about ChatGPT account data, OpenAI retention, model training or platform privacy controls should be directed to OpenAI through its current privacy channels.
Tambition may update this notice when the GPT configuration, website, knowledge workflow or external integrations materially change. A revised date should be shown when substantive changes are made.
Use z/OS Agent Elephant Safely
Sanitized evidence in. Human-reviewed decisions out.
This notice is an operational privacy and responsible-use statement. It does not replace your organization’s confidentiality, security, legal, regulatory or change-management requirements.