Sports Veritas operates under a single mandatory rule: no article reaches publication without human editorial approval and byline ownership.

This design choice aligns directly with the human-in-the-loop requirements of the EU AI Act. It is not a coincidence.

What the EU AI Act requires

The EU AI Act defines high-risk AI systems as those that pose identifiable risks to health, safety, fundamental rights, or the environment. High-risk systems must include meaningful human oversight — not rubber-stamping, but actual human control at critical decision points. The AI contributes; the human decides.

Articles about championship achievements meet this threshold. An achievement record affects an athlete's credentials, their qualification pathways, and their international record. A factual error or misattribution in publication is consequential.

How Sports Veritas implements it

The automated drafting system on Sports Veritas is built to support human decision-making, not replace it.

A draft article combines the achievement data, athlete commentary, official quotes, federation context, and results data. But the draft is not the article — it is input to the reviewer. The reviewer reads the draft, checks the facts, verifies the attributions, confirms the photo credits, and decides whether to publish. Only then does the article go live.

The reviewer's name appears as the byline. This is not decoration. It is accountability. The reviewer stakes their reputation on the article.

If a draft cannot be corrected or approved, the reviewer rejects it. The achievement declaration remains public; only the article is not published. That separation is deliberate and enforced by the platform, not by policy.

What this excludes

Meaningful human oversight is not:

• Automated content moderation that a human acknowledges.

• A checkbox that a staff member clicks before publishing.

• A notification that a human receives after an AI system has already decided.

Human-in-the-loop requires that the human intervention is the point where consequential decisions are actually made.

Why this matters for federations

A federation's reputation depends on the authority of its records. An athlete who wins a world championship and sees their achievement published with misattributed quotes or incorrect data has legitimate grounds to distrust the federation's record-keeping.

The EU AI Act recognises this. Where AI systems handle high-consequence outputs, humans must remain in control. That is not a burden imposed by regulation. It is what good record-keeping requires.

Sports Veritas is built that way because it is the right way.

NxtStride Finland Oy builds digital recognition infrastructure for practitioner sports federations. The design of TitleStream and Sports Veritas reflects commitments to human oversight that precede and exceed current regulatory requirements.