AI & Cybersecurity Results Disclaimer

1. Decision support, not a guarantee

PhiShark results are signals and recommendations for qualified human review. They are not a certification that content is safe or malicious and are not legal, compliance, financial or incident-response advice.

2. Known limitations

• False positives, false negatives, stale intelligence, unavailable sources and incomplete scans may occur.
• Attackers may evade, poison or manipulate automated, AI and prompt-based analysis.
• A score can change as content, infrastructure, models, policies or threat intelligence changes.
• File, QR, URL, email and webpage inspection may not observe behavior limited by credentials, geography, timing, sandbox detection or later modification.

3. User responsibilities

Independently verify high-impact results; maintain layered controls, backups, patching, access management and incident response; restrict submissions to authorized data; and escalate suspected incidents to qualified personnel. Do not automatically block, disclose, accuse or discipline a person solely because of a PhiShark result.

4. No complete-security promise

PhiShark does not promise to prevent every attack, compromise, harmful message, malicious instruction, business interruption or data loss. Contractual service levels and mandatory liability rules, if applicable, remain unaffected.

5. Suspected errors and incidents

If a result appears wrong or an incident is suspected, preserve evidence, isolate affected systems where appropriate and contact [email protected]. Do not send credentials, full payment data or unnecessary sensitive personal data in the report.