Responsible AI Policy
Version: 1.0
Status: Published
Effective Date: July 2026
Next Review: July 2027
1. Our Commitment
Artificial Intelligence (AI) is an important part of how we work at Curia AI. It helps us research, analyse information, develop ideas, draft documents, build software and improve efficiency.
AI does not replace our professional judgement or accountability. Every piece of advice, recommendation and client deliverable remains the responsibility of a human consultant.
We believe organisations should understand how AI is used on their behalf. This policy explains how we use AI responsibly within our own business and when delivering services to our clients.
2. Our Principles
Everything we do with AI is guided by six principles.
Human judgement comes first
AI supports our work but does not replace human expertise. Every significant output is reviewed, challenged and approved by a suitably qualified consultant before it is shared with a client or published.
Privacy and confidentiality matter
We apply the same standards of confidentiality and data protection when using AI as we do throughout the rest of our business. We minimise the use of personal data wherever possible and only use AI in ways that are consistent with our legal and professional obligations.
Transparency builds trust
We are open about when AI has played a material role in producing work. Where appropriate, we will explain how AI has been used, what safeguards have been applied and where human judgement has been exercised.
We use AI responsibly
We only use AI in ways that we would recommend to our own clients. That means considering governance, privacy, security, fairness, environmental impact and human oversight before adopting new technologies or new ways of working.
We also seek to use AI efficiently and proportionately. This includes selecting the most appropriate technology for each task, designing workflows that minimise unnecessary processing and using computational resources responsibly. Efficient use of AI helps us deliver value to clients while reducing unnecessary cost and environmental impact.
We design governed workflows
Responsible AI is not simply about choosing the right technology. It is about designing effective workflows. Our processes define where AI contributes, where human review takes place, where evidence is verified and who remains accountable for the final outcome.
Accountability never changes
Using AI never transfers responsibility. Every client deliverable remains owned and approved by Curia AI, regardless of the role AI played in producing it.
3. How We Use AI
AI may assist us with activities including:
- Researching and synthesising publicly available information.
- Analysing documents and identifying themes.
- Drafting reports, proposals and presentations.
- Developing software, prototypes and technical solutions.
- Supporting coding, testing and documentation.
- Creating internal tools and workflows.
- Improving productivity and administrative efficiency.
AI helps us work more efficiently, allowing us to spend more time applying professional judgement, critical thinking and experience to the work we deliver.
4. How We Do Not Use AI
We do not use AI to:
- Make solely automated decisions about individuals.
- Replace professional judgement or accountability.
- Create misleading, deceptive or fictional content.
- Generate fake testimonials, case studies or evidence.
- Process sensitive personal data without appropriate safeguards.
- Circumvent legal, regulatory or ethical obligations.
5. Human Oversight
Every significant client deliverable that has been produced with AI assistance is reviewed before it is shared.
Human review means more than checking grammar or formatting. It includes:
- Reviewing accuracy.
- Challenging assumptions.
- Verifying important facts and sources.
- Considering context and proportionality.
- Applying professional judgement.
- Confirming the work is suitable for its intended purpose.
No significant client recommendation is delivered without meaningful human oversight.
6. Protecting Information
Protecting client information is fundamental to the way we work.
Before using AI we consider:
- Whether AI is appropriate for the task.
- The sensitivity of the information involved.
- Whether personal data can be removed or anonymised.
- Whether the technology provider meets our privacy and security requirements.
- Whether additional governance measures or risk assessments are required.
Where processing presents a higher level of risk, we apply additional governance controls before AI is used.
7. Using the Right AI for the Right Task
Different AI systems have different strengths. Rather than relying on a single provider, we select the technology that is best suited to the task, taking account of reasoning capability, coding performance, research support, privacy commitments and governance controls.
8. Governance and Continuous Improvement
AI technology continues to evolve rapidly.
We regularly review:
- The AI tools we use.
- Developments in regulation and guidance.
- Emerging good practice.
- Our own internal processes and governance arrangements.
Where necessary, we update our ways of working to ensure they remain proportionate, transparent and aligned with current best practice.
9. Learn More
If you would like to understand more about how AI has been used in a particular piece of work, or would like further information about our approach to AI, we would be pleased to discuss it with you.
Our detailed internal AI Governance Policy is available to clients and prospective clients during procurement, due diligence or on request.
For general enquiries, please contact hello@curiaai.co.uk.
Responsible AI with Purpose
At Curia AI, we believe AI should help organisations achieve more for the people and causes they serve.
Responsible AI is not about slowing innovation or adding unnecessary process. It is about creating the confidence to adopt AI well, the capability to use it effectively, and the governance to ensure it remains trusted as organisations grow.
That philosophy shapes how we use AI within Curia AI and how we help our clients realise the benefits of AI responsibly, sustainably and with confidence.
This policy is reviewed at least annually and whenever there is a material change to the AI technologies we use or the services we provide.