AI transparency statement

ClubTitan uses AI to take admin off volunteers' plates. No club management platform we know of publishes a statement like this, but parents and volunteers deserve to know exactly what the AI does — and, more importantly, what it can never do. Everything on this page describes how the system is actually built, and it forms part of the club terms and member terms.

What the AI does

• drafts things a person then reviews — match reports from a coach's voice note, newsletters, replies to routine messages;
• polishes writing a person has already written;
• reads routine content so a volunteer doesn't have to — parsing availability replies, reading receipts for expense claims;
• answers members' questions about their own club — training times, what's owed — using only data that member is already allowed to see;
• triages inbound messages: the most it can ever do with one is draft a reply for a human to approve, or flag it to a human.

The human-approval rule

Every AI draft is reviewed and approved by a person before anything is published or sent. AI output re-enters the platform through the same doors a human's work does — the same permission checks, the same audit trail — and nothing leaves the platform (a message, a post, an email) without a person choosing to send it.

What the AI never does

• never publishes or sends anything on its own;
• never moves, collects or refunds money;
• never makes safeguarding decisions — those always belong to people;
• never makes a decision with legal or similarly significant effects about you by machine alone (the UK GDPR Article 22 standard) — routing and reminders are ordinary deterministic software, and judgements are human.

Children's data and AI

The most sensitive categories of data on the platform — safeguarding concerns, medical detail and the like — are classified in our systems and excluded from AI prompts by that classification, not by policy alone. Children are never profiled by AI, and AI is never used to market to anyone under 18.

Cost caps

Every club's AI usage is metered and capped, so AI features can never run away with a club's money — core AI features are included free within a generous cap, and premium AI is priced in the fees schedule.

Who processes AI requests

Our AI model provider is Anthropic, listed with the rest of our sub-processors. Anthropic's commercial API terms provide that content submitted through its API is not used to train Anthropic's models; API inputs and outputs are retained by Anthropic for a limited period for safety and abuse monitoring, then deleted [both statements to be re-verified against Anthropic's current commercial terms before launch]. Where a club brings its own AI provider key, that usage runs on the club's own relationship with its chosen provider, under that provider's terms.

Is AI-drafted content labelled?

The person approving a draft always knows it was AI-drafted. Whether published content also carries a reader-facing label is a decision we have not yet made — the posture will be settled and stated here before launch [to be confirmed before launch].

Connecting your own AI assistant

Members can point their own ChatGPT, Claude, Copilot or Gemini at their club data. That connection is read-only, scoped to exactly what the member is already allowed to see in the app, authorised by the member signing in, and revocable by the member at any time from Settings. What you and your assistant say to each other happens in your assistant, under your AI provider's own terms — ClubTitan only answers the read-only questions your assistant asks on your behalf.

This draft was last updated in July 2026.