AI Policy
OUR COMMITMENT IN ONE LINE
We build AI for regulated industries, with a human in the loop, clear disclosure when AI is used, and your data staying under your control. That is the starting brief on every build.
1. Who we are and what this covers
Clone Centre Ltd. ("we", "us") builds AI software for regulated industries. This policy explains how we design, build, and operate our products, TRAICE™ (AI governance) and ANSAIR™ (AI communications), in line with the responsible-AI obligations that apply to us and to our customers.
Our customers are often deployers of AI themselves: law firms, financial services, healthcare, and government bodies. We build the controls they need to meet their duties too.
2. The frameworks we align to
European Union: the EU AI Act
Regulation (EU) 2024/1689, the EU AI Act. Our products are designed to support the obligations most relevant to our customers, including:
- Article 4, AI literacy. We equip our staff and, through in-product guidance, our users with the understanding needed to operate AI systems responsibly.
- Article 14, Human oversight. Our systems are built to be overseen, interpreted, overridden, and interrupted by a person. No product runs unsupervised on consequential decisions.
- Article 50, Transparency. Where people interact with our AI (voice, chat, automated messages), we support disclosure that they are engaging with an AI system.
- Article 73, Transparency and technical documentation. TRAICE™ exists to make AI usage traceable: who used it, on what matter, and what changed.
United Kingdom: the UK AI framework
The UK takes a principles-based, regulator-led approach. Our products are designed against the five principles set out in A pro-innovation approach to AI regulation:
- Safety, security and robustness. We test before we ship and monitor in production.
- Transparency and explainability. We disclose AI use and keep records of what AI did.
- Fairness. We design against bias and monitor for disparate impact.
- Accountability and governance. A named human is accountable for every product's outputs.
- Contestability and redress. People can query, challenge, and escalate AI decisions to a human.
Data protection
Where personal data is processed, our products are designed to be compatible with the UK GDPR and EU GDPR: data minimisation, purpose limitation, and the right for individuals to understand and contest automated processing.
3. How this works in practice
Human-in-the-loop by default
Our AI assists; it doesn't replace judgement. Outputs that could meaningfully affect a person, like a legal position, a benefit decision, or an emergency referral, are surfaced for human review before they take effect.
Data sovereignty
Our customers' data stays under their control. We do not sell personal data, and we minimise what we process to what each job strictly requires.
Transparency to end-users
When someone speaks to ANSAIR™ on the phone or reads AI-assisted work in TRAICE™, they should know AI was involved. Disclosure is built in from the start.
Audit trails
TRAICE™ records prompts, outputs, edits, and the human decisions made on top of them, so usage can be reviewed, reported, and shown to a regulator if asked.
Proportionality
Higher-risk uses get more controls. We do not treat a low-stakes automation the same as a tool influencing a legal, financial, or clinical decision: the safeguards scale to the stakes.
4. What we don't do
- We don't build social scoring.
- We don't build untargeted facial recognition.
- We don't sell personal data, ever.
- We don't deploy AI for consequential decisions about people without a human able to step in.
5. Roles and accountability
Within Clone Centre, every product has a named owner accountable for its responsible-AI posture. Our customers, as deployers, remain accountable for how they configure and use our products in their own context, and we give them the controls to do that well.
6. Incident reporting
If something goes wrong, a harmful output, a transparency failure, a data issue, we want to know fast. Report concerns to safety@clonecentre.ai and we will investigate and act.
7. Keeping this current
AI regulation is moving. The EU AI Act's deadlines continue through 2027 to 2028, and the UK framework is evolving. We review this policy as the law and our products develop, and we'll show the "last updated" date honestly above.
QUESTIONS?
For anything about this policy, email hello@clonecentre.ai. For a specific product, see its own terms within the app.