Our Approach to Artificial Intelligence

Mid Devon District Council is committed to using technology responsibly to improve services for residents, businesses, councillors and staff.

Artificial Intelligence (AI) has the potential to help the Council work more efficiently by reducing administrative tasks, improving access to information and supporting officers in their day-to-day work. However, AI must always be used safely, transparently and ethically.

The Council's use of AI is governed by formal policies, information governance controls, data protection requirements and oversight arrangements to ensure that residents' information remains protected and decisions remain accountable to people, not machines.

What is Artificial Intelligence?

Artificial Intelligence (AI) is technology that can assist with tasks that would normally require human input. Examples include:

• Drafting documents and correspondence
• Summarising lengthy reports and consultations
• Research and information retrieval
• Data analysis and trend identification
• Knowledge management and information discovery
• Supporting administrative activities

AI is designed to support staff and improve productivity. It does not replace professional judgement, officer accountability or democratic decision-making.

How Mid Devon Uses AI

The Council currently uses and evaluates AI technologies to support productivity, innovation and service improvement.

Examples of approved or explored uses include:

• Preparing draft reports, briefings and correspondence
• Summarising information and consultations
• Supporting research and information gathering
• Improving access to organisational knowledge
• Assisting with administrative processes
• Exploring specialist use cases such as Local Plan production and data analysis

Any use of AI must be supported by appropriate governance, information security and data protection controls.

Human Oversight and Accountability

The Council does not delegate accountability to AI systems.

Where AI is used:

• Officers remain responsible for all decisions.
• AI-generated content must be reviewed and validated by a person before use.
• Published information remains the responsibility of the Council.
• Residents can request clarification where AI has supported a process.
• Significant decisions affecting individuals are not made solely by AI.

Human judgement, professional expertise and democratic accountability remain central to Council services.

Protecting Your Information

The Council recognises the importance of privacy and data protection.

When considering AI technologies, the Council will:

• Assess information governance risks.
• Complete Data Protection Impact Assessments where required.
• Ensure compliance with UK GDPR and data protection legislation.
• Restrict the use of personal or sensitive information in unauthorised AI tools.
• Apply appropriate security, audit and access controls.

The Council's preferred AI platform is Microsoft Copilot because information remains within the Council's controlled Microsoft environment and is protected by existing security and governance arrangements.

Fairness, Equality and Transparency

The Council is committed to ensuring AI is used fairly and without discrimination.

When introducing new AI-enabled services, the Council will consider:

• Equality impacts
• Potential bias
• Accessibility requirements
• Security risks
• Legal and ethical implications
• Environmental considerations

The Council is committed to being transparent about how AI is used and will continue to review emerging guidance and best practice from the public sector.

Residents have rights in relation to automated decision making.

Where a decision could have a legal or significant effect on an individual, the Council will ensure appropriate safeguards are in place, including:

• Human review where required
• The ability to challenge decisions
• Compliance with data protection legislation
• Transparency about how decisions are reached

The Council maintains oversight of systems that contain elements of automated decision making.

Using AI When Contacting the Council

You may choose to use AI tools to help draft complaints, representations, appeals or correspondence to the Council.

If you do:

• Read the content carefully before submitting it.
• Make sure it accurately reflects your own views and experiences.
• Avoid asking AI tools to add unnecessary legal language or complex references.
• Do not share sensitive personal information with public AI systems unless you understand how it will be used.
• Keep submissions clear and concise.

We gain the best understanding of your circumstances when information is presented clearly and accurately.

Governance and Further Information

The Council maintains:

• An Artificial Intelligence Policy
• Data Protection and Information Security policies
• AI governance arrangements
• Risk assessment and approval processes for new AI solutions
• Training and awareness programmes for staff

These arrangements help ensure AI is adopted responsibly, transparently and in a way that benefits residents while protecting public trust.

Contact Us

If you have questions about the Council's use of AI, information governance, privacy or data protection, please contact the Information Management Team via dpo@middevon.gov.uk

Using AI to contact us or make a representation

We want to hear your feedback or representations using your own words.  You may want to prepare your submission with the help of Artificial Intelligence (AI) but these tools can sometimes make submissions too complex or include legal references that are irrelevant or misleading.  We gain the best understanding of your submission when it is raised clearly in your own words. 

If you do want to use AI to support you, please consider the following:

• It is your submission so make sure you read it through carefully to make sure it is accurate before you send it to us.

• Focus on your experience and the impact it had on you.
• Do not ask AI to add legal arguments or complex references.
• Be careful using general AI tools and do not share personal or sensitive information which might be used to train that AI system and/or be processed in a different country.
• You are responsible for ensuring that any AI-generated content submitted is relevant, accurate and as short as possible to prevent delays.

• Where we feel that your use of AI is negatively impacting on our understanding of your submission, or is overly long or complex, you may be asked to resubmit or approve a more concise statement.