WOMEN4AI CALL TO ACTION 2022
DOWNLOAD HERE
PREAMBLE
This call to action invites organisations to commit to implement inclusive AI, whether that means embarking on first steps, accelerating existing initiatives or scaling best practices. To create inclusive technology, fundamental transformation is needed: this includes supportive governance frameworks and structural shifts, but small steps are a good way to start this journey.
Actions of all sizes to tackle bias and embed inclusive practices will close gaps, bringing the world closer to an inclusive AI ecosystem. As such, organisations can contribute to systemic change in the broader ecosystem through internal transformation.
As AI reflects human behaviour, it is essential to tackle bias not just in AI but also in society to create truly sustainable inclusion.
We define inclusive AI as a subfield of ethical AI which accounts for needs of different groups, including minorities, marginalised and underrepresented groups.
The application of inclusive AI addresses the issue of bias and discrimination with the aim of reducing inequalities, including representation, accessibility and interpretability. Inclusive AI is non-discriminatory in its production, unbiased in its consequences, and accessible to all.
VISION STATEMENT
When AI is developed and deployed in an inclusive manner, it will drive better outcomes for women and the world.
Our vision statement can be achieved through the following commitments, which are divided into 3 themes.
OUR COMMITMENTS
We ask organisations to choose 2 commitments at a minimum; these can be selected from the same theme, or from different themes.
1. Strengthen your organisation’s pipeline and practices
-
We commit to fostering a supportive ecosystem for women and other marginalised people, including racial and ethnic minorities in tech roles through targeted policies within our governance structures.
-
We commit to providing training and capacity building programmes to mainstream inclusive AI adoption within our organisation.
2. Build sound data foundations
- We commit to using relevant approaches to identify and flag incomplete or unrepresentative data and to mitigate bias within our AI datasets and development programmes.
3. Engage with your suppliers
- We commit to using relevant approaches to identify and flag incomplete or unrepresentative data and to mitigate bias within our AI datasets and development programmes.
- We commit to implementing a capacity building programme to raise awareness, share knowledge and best practices with suppliers on identifying and avoiding non-Inclusive AI.
Overall targets:
We ask organisations to commit to one target from the following targets:
- We will share examples of good practices or case-studies that show progress towards Inclusive AI.
- We will complete and share an Algorithmic Impact Assessment by November 2023.
- Additional resources: W4AI Impact Assessment Tool in the Inclusive AI Toolkit.