Legal & Trust

AI Safety

Last updated: 2026-07-29

SolVida (part of OptFin) builds AIs to help people, not to replace their judgment. This page explains the principles and safeguards behind Luna and the SolVida AIs.

AIs assist humans, with human oversight

Luna and the SolVida AIs are tools that assist you. Work stays under human oversight: people remain in control of decisions, and higher-risk actions are designed to require human approval rather than running unchecked.

A blocking compliance gate

Client-facing output passes through a blocking compliance gate before it is sent. This gate is a safeguard on what leaves the platform and helps prevent inappropriate or non-compliant messages from reaching people.

No fabricated guarantees or advice

Our AIs are built to avoid fabricating guarantees, credentials, or facts. AI output is assistance, not professional, financial, or legal advice, and it does not promise specific outcomes.

Honest documents: restored, not faked

When SolVida improves a document (for example a scan or photo), it restores it to a legible, lightweight version and preserves your original as evidence. We do not fabricate content that was not in your document.

Limitations — please verify

AI systems can be wrong, incomplete, or out of date. You should review and verify important outputs before relying on or acting on them, and consult a qualified professional for regulated matters (legal, financial, tax, medical, and similar).

Report a concern

If you see AI behavior that seems unsafe, misleading, or harmful, please tell us at jose@pmafin.com so we can review it.

We describe safeguards that are actually in place today and will update this page as they evolve. See also our Security and Terms of Service pages.

Contact

SolVida (part of OptFin). jose@pmafin.com.