Session organized by the United Nations Development Programme (UNDP)
Session Description:New technologies rely heavily on artificial intelligence system – a system that operates and learns from the data. Both the source and the end consumer of this data are human beings. And when that data is generated by and collected via humans, it carries all the biases we do, including biases about women. Algorithmic bias comes from simply uploading it along with everything else when we use machine learning and can amplify and perpetuate gender bias by uploading unrepresentative data sets. Reliance on underlying language processing and algorithms have demonstrated gender bias in the context of employment advertising and recruitment tools. For example, NLP, a critical ingredient of standard AI systems like Amazon's Alexa and Apple's Siri, among others, has been
found to show gender biases. Similarly, facial recognition technologies have been called out for
disproportionately misidentifying women, particularly women of colour. As new technologies expand at an unprecedented pace, it is crucial to evaluate how they depict and reinforce existing gender bias and stereotypes.
Session Objectives:
The key objectives of this session are to:
- Understand the gender biases in new technology – causes, forms and impact;
- Assess the most pressing gaps in technology governance that still enable gender bias and reinforce inequality;
- Explore notable (if any) policy or regulatory changes have been made to mitigate algorithmic bias;
- Identify ways to address gender bias in Artificial intelligence.
Key questions:
Panellists will reflect on the following questions:
- How do new technologies depict and reinforce existing gender bias and stereotypes, and how their development teams' composition affects these portrayals?
- What are the most significant gaps in technology governance that perpetuate and exacerbate gender bias and stereotypes?
- Have there been any examples where governments have taken note of gender bias and stereotypes in new technologies and addressed them through policy and regulation?
- What are some of the efforts that businesses can adopt/have adopted to altogether avoid the larger structural and systemic forms of bias and discrimination perpetuated and amplified through algorithmic technology?
Background to the discussion:
The fourth Industrial revolution (4IR) represents a new era that builds and extends the impact of digitization in new and unanticipated ways. The 4IR is driven by new technologies, such as artificial intelligence, augmented reality, robotics, neurolinguistic programming (NLP), sentiment analysis and 3-D printing. These emerging new technologies are increasingly influencing people's opinions and behaviour in everyday life - changing the way humans create, exchange, and distribute value. It was anticipated that these new technologies would rebalance our societies' inequity, moving toward a more inclusive, human-centred future together. 4IR is still evolving, and it's impacts are yet to be seen; however, it is argued that the 4IR will exacerbate inequality, threaten security and risk identity, voice and community.
