Skip to content

AI & Reliability Disclaimer

Last updated: June 2026

Please read this before you build with OIOXO. It explains, in plain words, what OIOXO can and can’t do, and what you’re responsible for. It is part of our Terms of Service; where the two differ, the Terms control.

AI is imperfect

OIOXO uses artificial intelligence to write, change, and run code. AI is probabilistic and inherently imperfect. Its output can be wrong, incomplete, insecure, inefficient, out of date, or simply not do what you asked — even when it looks confident and correct. Treat everything it produces as a draft to review, not a finished answer to trust.

The on-device models are small — and that’s the point

OIOXO’s defining feature is that its models run on your own device, privately, for free. To do that, they have to be small. Small models are less capable than the large cloud models you may be used to, and they’re more likely to struggle with complex tasks and to produce lower-quality or incorrect results. This is a deliberate trade-off for privacy and cost — not a defect. For harder work, you can bring your own API key to use a stronger model.

“Verified” doesn’t mean “correct”

OIOXO checks generated code by running it — compiling it and executing tests where it can. This catches many mistakes, which is why we built it. But it cannot catch everything. A build that passes can still be wrong, insecure, or unsuitable for your purpose. A green result is a good sign, not a guarantee.

Always review and test before you rely on it

You are responsible for reviewing, testing, and validating anything OIOXO produces before you use it. Don’t deploy AI-generated code to production, handle sensitive or regulated data with it, or use it where a mistake could cause harm or loss without independent review by a qualified person. What you ship is your responsibility.

Keep backups of your work

OIOXO is an editor: its AI agents can create, change, overwrite, and delete files in the projects you open, and software can fail. OIOXO snapshots changes where it can so edits are undoable, but that is not a substitute for your own backups. Use version control (commit your work to Git) and keep copies of anything important — especially before letting an agent make large changes. We are not responsible for loss of or damage to your code, data, or projects.

Provided “as is”

OIOXO, its models, and its output are provided “as is” and “as available”, without warranties of any kind. To the maximum extent permitted by law, OIOXO is not liable for losses arising from your use of the Service or its AI output, including lost code, data, time, profits, or business. The full warranty disclaimer and limitation of liability are in the Terms of Service.

In short

  • AI makes mistakes — review everything before you trust it.
  • On-device models are smaller and weaker than big cloud models.
  • “Verified” is a strong signal, not a guarantee of correctness.
  • Keep your own backups and use version control.
  • You’re responsible for what you build and ship; OIOXO is provided as is.

Questions? support@oioxo.com.