Please note that this work is still in the first stage and additional work is ongoing. For information on who has participated in this work so far, and information on how to get involved, please see our working group page.
Introduction
In the fast and ever changing and moving world of Artificial Intelligence there are a growing number of products which are being offered to social care commissioners and service providers.
Supported by the Oxford Institute for Ethics in Artificial Intelligence a small group of interested individuals have met on several occasions to draw up a set of principles which we believe will be useful for individuals and organisations to consider when either commissioning, purchasing or using AI tools. We have sought to ensure that these principles are sensitive to the experience of those who use services and support.
We acknowledge that these principles will develop over time. They are not the last word, but we hope they will provide a short and accessible guide for use.
We believe the following principles should be considered:
| Principle | Other adjacent principles that may be linked |
|---|---|
| Truth | Transparency, Validity, Explicit |
| Transparency | Honesty, Justification, Explainability |
| Equity | Fairness, Non-discrimination, Human Rights, Inclusivity |
| Trust | Privacy, Accountability |
| Accessibility | Non-discrimination, Equality |
| Humanity | Freedom, Person-led, Empathy, Human oversight |
| Responsiveness | Review, Assessment, Evaluation |
Putting principles into practice
In the next section we will explore what we mean by these principles and offer some practical scenarios in which they may be considered. We will do so by using a human rights tool to explore ethical decision-making in Ai use.
The FAIR human rights model, as developed by the Scottish Human Rights Commission, Facts, Analysis of rights, Identification of responsibilities, and Review of actions—provides a clear and structured way to ensure that the rights and dignity of those in our care are protected when we introduce AI into the social care context.
In what follows we will consider how we can apply this framework to the key ethical principles that underpin the responsible use of AI – truth, transparency, equity, trust, accessibility, and humanity – and share some real-world examples of how these can play out in practice.
