Skip to content

Legal

Responsible AI

Last updated November 2026

This document describes how the platform behaves. Have it reviewed by qualified counsel for your jurisdictions before launch.

AI runs throughout SkillForge, under clear rules.

  • Consent first. AI features stay off until you turn them on, and you can turn them off at any time.
  • Data minimisation. Requests carry only what the task needs. Email addresses and phone numbers are removed before anything leaves SkillForge, and your name and identifiers are never sent.
  • Grounded answers. The learning assistant answers from your course material first and labels anything beyond it as general explanation.
  • No fabrication. AI never produces grades, certificates, qualifications or institutional records; those come only from official data.
  • Humans decide. Job matching explains its reasoning to people who make the decisions. AI does not make hiring decisions.
  • Provider independence. SkillForge can use different AI providers. If none is available, features fall back to built-in, rule-based guidance, clearly labelled.
  • Oversight. Administrators monitor usage, costs and errors, and every request is logged without its content.