The fine print.

Short version: this is a toy. Do not let anything real depend on it. Last updated September 21, 2026.

1. What this is

heyjev.ai is an independent, experimental website built for fun and to show how typed AI decisions work. It is not an official product of TypeSafe AI or of any other company. “Hey Jev, should I deploy?” is a deployment vibe check: a joke with real probabilities behind it.

2. Do not rely on it. Really.

The verdicts, reasons and probabilities are entertainment and education. They are not advice, not a risk assessment, not a deployment gate and not a substitute for tests, reviews, rollback plans or your own judgment. Do not use the site or the API to decide, approve, block or automate a real release, or any other decision with consequences. If you do, you do so entirely at your own risk and responsibility, and you agree that the operator has no part in that decision.

3. Nothing depends on this

There is no service level, no uptime commitment and no support. The API, its response format, the questions, the verdicts, the limits and the site itself may change, break or disappear at any time without notice. Do not build products, pipelines, integrations or workflows that depend on heyjev.ai being available or behaving the same way tomorrow. If you wire it into something anyway, that something is yours to fix when it stops working.

4. Permitted use

You may use the site and its API for personal, non-commercial purposes: trying scenarios, sharing verdicts, learning how typed AI decisions work, and calling the API from your own tools within the published limits, always subject to sections 2 and 3.

5. Commercial use requires authorization

Any commercial use of heyjev.ai, its API, its verdicts, its question set or its output is prohibited without explicit written authorization from the operator. Commercial use includes, without limitation: embedding the service in a product or service that is sold or monetized, reselling or sublicensing access, bulk or automated consumption beyond the published rate limits, using the output to train or evaluate commercial models, and presenting the service or its output as your own. Ask first; the answer may well be yes.

6. Fair use of the service

Do not attempt to circumvent rate limits or quotas, probe or attack the infrastructure, scrape at scale, or send content you do not have the right to share. The operator may throttle, block or suspend access at any time.

7. What happens to what you type

The scenario you type is sent to TypeSafe AI’s API to be evaluated and may be cached in memory for a short time to serve identical requests. Do not include secrets, credentials or personal data. Share links encode the scenario text in the URL, so anyone with the link can see it. The site may collect anonymous usage analytics.

8. No warranty, no liability

The service is provided “as is” and “as available”, without warranties of any kind, express or implied, including accuracy, fitness for a particular purpose and availability. Probabilities are model judgments about the text you wrote, not statements about your systems. To the maximum extent permitted by law, the operator is not liable for any direct, indirect, incidental or consequential damages arising from the use of, reliance on, or unavailability of the service, including any deploy you do or do not make and anything that breaks as a result. You assume all risk.

9. Third parties

Jev is a model by TypeSafe AI, used through its public API under its own terms. The site is hosted on nullplatform. Their names and marks belong to their owners; their appearance here does not imply endorsement.

10. Changes

These terms may change at any time. The current version is always at this address. Continued use means you accept the changes.

11. Contact

Authorization requests and questions: @geisbruch on X.