Ask MORIS · from The Resonance Institute
Bring something you are actually wrestling with. See what an LLM with an artificial conscience makes of it.
This is a private beta, provided as is. Answers here are generated automatically and may be inaccurate.
Nothing here is medical, psychological, legal, or financial advice, and it is not a substitute for a professional. The Resonance Institute, LLC accepts no responsibility or liability for anything you read here, or for any decision you make.
Do not enter anything you would not want a third party service to receive.
This page is not monitored. If you are in crisis, please reach a person. In the United States, call or text 988. Elsewhere, contact your local emergency number.
What you have written sounds like more than this page should answer. It has not been sent anywhere and nothing has been judged.
In the United States, call or text 988 for the Suicide & Crisis Lifeline, or text HOME to 741741. Elsewhere, contact your local emergency number or crisis line.
If you are in immediate danger, please call your emergency number now.
Sign in for 10 turns a day on the free tier, counted on your account rather than this browser. Anonymous use stays as it is.
Deleting your account removes the account and its counters. Anonymous counts stay. The thread and the sign-in kept in this browser are cleared with it. This cannot be undone.
Output is generated automatically and may be inaccurate. Not advice of any kind, and no liability is accepted for it. This page is not monitored. If you are in crisis, please reach a person: in the US call or text 988, or your local emergency number.
Anonymous use is limited to 10 exchanges a day in this browser. Accounts and higher limits are coming.
Or try one of these:
. The packet appears verbatim inside the prompt that was
delivered, so the text above is what the answering model received, not a description of it
written afterwards.
full prompt — ·
packet within it —
These two are digests of different things and are not expected to be equal: the first covers
the whole delivered prompt, the second the conscience packet carried inside it.
The judgment is mechanical, not generated. No model is used in the judgment, or in producing what MORIS sends to the answering model. The same reading always produces the same verdict and the same weight.
The reading itself is a model's, and it is not perfectly consistent: across the paper's 680 MoralChoice dilemmas, each read twice, the same model returned the identical nine fields on 83.3% of comparable readings. That is a property of that text as much as of the reader. No other corpus has been measured under this reader, so no other number is shown.
The model that answers you is still itself: its own voice, its own judgment, its own safety filters. What has changed is that its answer is bound to align with MORIS.