AI can be wrong
Generated content may be incomplete, inaccurate or unsuitable for your circumstances. Treat it as material for reflection, not a ruling on fact.
Safety, AI boundaries and user protection
Symbols can open understanding, but they cannot replace reality
REGENESIS is designed for symbolic reflection and record-keeping. You retain interpretation, real-world judgment and the right to seek professional help.
Core boundaries
Safety begins with an honest product definition
Not medicine or therapy
REGENESIS does not diagnose, prescribe treatment or replace physicians, therapists or other qualified professionals.
Not legal or financial advice
Do not rely on product output for legal, investment, debt, insurance or other high-stakes financial decisions.
No harmful certainty
The product should not make definitive claims about death, disaster, illness, pregnancy, crime or relationship facts, or encourage fear, dependence or harm.
Planned release architecture
Safety mechanisms will continue to evolve with the service
The principles below describe plans for the formal service; they do not claim that a live safety center is already operating.
Rules before and during release
REGENESIS’s release architecture will use content boundaries, restrictions for high-risk topics, clear disclaimers and escalation paths where needed. In the formal service, we plan to record reproducible safety issues and assess their severity.
Correction, withdrawal and notice
When a class of result is confirmed to create serious risk of harm or material deception, we plan to correct rules, restrict further generation and, where reasonably feasible and contact is possible, provide affected users with withdrawal or correction notice. Product, technical and legal review will define the final mechanism before release.