AI is transforming many sectors of the economy and our day-today lives. We hear of success stories including medical advances, but also worries that AI will destroy jobs or even be an existential threat to humanity. We also know that for previous waves of technology – mechanical, electronic and digital – the costs and benefits do not fall equally to everyone in society. There are clear dangers that AI will further entrench existing power and deepen the digital divide: both at an individual level and globally. For example, training foundation models, such as GPT-4, requires enormous computational power and massive data sets accessible only to the largest corporations. However, the ways in which these can be used generatively in more niche areas, offers potential for minority languages and individualised learning that was previously only accessible to the rich. Whether the threats of AI or its opportunities dominate is not simply an abstract question, but one that impacts the most disadvantaged around us, and one that, as researchers and practitioners in digital technology, we can affect. If we truly want an inclusive digital society, then we need to make it happen.
Slides
Video
Microsite on AI for Social Justice Clara Crivellaro and Alan Dix (2021). AI and Social Justice: From Avoiding Harms to Positive Action including full notes of the talk (PDF, 2.8Mb) Video of talk at AI Summit London 2021 Download notes of the AI Summit talk (PDF, 2.4M) |
Digital economy and social justice
[zoom image]
What AI does
[zoom image]
How AI changes Society
[zoom image]
Why AI now?
[zoom image]
Network effects
[zoom image]
Peacock tails and runaway AI
[zoom image]
Signs of hope
[zoom image]
Reducing the cost of AI
[zoom image]