UnBias Fairness Toolkit

The UnBias Fairness Toolkit is now available to download and use. It aims to promote awareness and to stimulate a public civic dialogue about algorithms, trust, bias and fairness. In particular, on how algorithms shape online experiences, influencing our everyday lives, and to reflect on how we want our future internet to be fair and free for all.

The tools not only encourage critical thinking, but civic thinking – supporting a more collective approach to imagining the future as a contrast to the individual atomising effect that such technologies often cause. The toolkit has been developed by Giles Lane, with illustrations by Alice Angus and Exercises devised by Alex Murdoch; alongside contributions from the UnBias team members and the input of young people and stakeholders.

The toolkit contains the following elements:

1. Handbook
2. Awareness Cards
3. TrustScape
4. MetaMap
5. Value Perception Worksheets

What the Toolkit Offers:

• A pragmatic and practical way to raise awareness and stimulate dialogue about bias, trust and fairness in algorithms and digital technologies.
• It is designed to make complex and often abstract ideas tangible and accessible to young people and to non-experts across society.
• It supports critical thinking skills that can help people feel empowered to make better informed choices and decisions about how they interact with algorithmic systems.
• It helps collect evidence of how people feel about the issues and what motivates them to share their concerns by contributing to a public civic dialogue.
• It provides a communication channel for stakeholders in industry, policy, regulation and civil society to respond to public concerns about these issues.
• It can also be used by developers of algorithms and digital systems to reflect on ethical issues and as a practical method for implementing Responsible Research and Innovation.

All components of Toolkit are freely available to download and print under Creative Commons license (CC BY-NC-SA 4.0).

Published by Proboscis. Funded through the EPSRC TIPS programme: EP/N02785X/1