Last updated: 7 October 2026
Ordis uses artificial intelligence. This page explains what that means in practice, so you
can use it safely. It is deliberately short.
1. You are talking to a machine
Ordis is an AI system, not a person. When Ordis answers you, that answer is
generated by a large language model — currently provided by DeepSeek — and is not written or
checked by a human before you see it.
2. AI output can be wrong
Large language models work by predicting likely text. That makes them useful, and it also
means they can:
- state things that are simply false, while sounding certain;
- invent facts, names, dates, quotations, citations or file contents;
- be out of date, because their training has a cut-off;
- miss context you assumed it understood;
- produce code or formulas that look right and do not work;
- misread a document, a screenshot, or a photograph.
This is a property of the technology, not a defect unique to Ordis.
3. What this means for you
Check anything that matters before you act on it. That is the whole rule.
Be especially careful with decisions that are hard to undo or that affect someone else —
legal, medical, financial, engineering, safety, or anything involving other people’s money,
health, or personal data. Ordis can help you think; it cannot take responsibility for the
decision.
4. Ordis can take real actions
Ordis is an agent, not only a chat window. When you ask it to, it can read
and write files, run commands, open applications, browse the web and send messages. Those
actions are real and can be destructive.
- Keep backups of anything important.
- Read what Ordis is about to do before approving an action that changes or deletes
something. - Start with low-risk tasks while you learn how it behaves.
5. Your prompts leave your machine
Ordis keeps your files, memory and session history on your own computer. But to generate a
reply, the text of your prompt is sent to a third-party AI provider. Do not put
passwords, API keys, private documents, or other people’s personal data into a prompt unless
you are comfortable with that. See our Privacy Policy.
6. AI-generated content on this website
Some text, code and images on this website and in our product were produced with the help of
AI, then reviewed by us. If you believe something here is inaccurate, tell us and we will fix
it.
7. If Ordis gets something wrong
Please tell us. Report it on any of our channels or email
zeratheonfoundation@gmail.com. During early
access, reports of wrong or unsafe output are the most useful thing you can send us — they are
how the system improves.