Achieving Independence from Tech Giants: One Person's Journey Towards a Morally Superior AI Era
In the rapidly evolving world of artificial intelligence (AI), ethical concerns have come to the forefront within tech corporations. This shift is exemplified by the high-profile departures of AI ethicists, such as Timnit Gebru and Meg Mitchell, from Google, and the subsequent establishment of the Distributed AI Research Institute (DAIR).
Founded by Timnit Gebru, a renowned AI researcher who parted ways with Google under controversial circumstances, DAIR was launched with a mission to counterbalance Big Tech’s dominant influence in AI research, development, and deployment. The institute aims to produce equitable, inclusive AI research that prioritises the needs of marginalised communities over corporate profit.
Timnit Gebru's vision for DAIR is deeply rooted in addressing the ethical challenges and harms embedded in AI technologies, particularly those perpetuated by large technology corporations that often prioritise profit over ethics and human well-being. Gebru and the DAIR initiative emphasise that AI development should be independent from profit-driven systems and inclusive of diverse perspectives.
The impact of DAIR is significant in promoting ethical AI that foregrounds equity, transparency, and community involvement. Unlike conventional research institutions focusing on producing academic papers accessible mainly to experts, DAIR aims to communicate its findings in ways that are accessible and useful to affected communities. This approach facilitates a more democratic and people-centered AI research process that proactively considers risks and harms from the outset rather than merely mitigating them post hoc.
DAIR operates remotely with teams located globally, reflecting its commitment to inclusivity and global perspectives. Its research projects are diverse and focused on using AI as a tool for social good, addressing real-world problems and prioritising community needs. For instance, Raesetje Sefala, a DAIR fellow, is working on a research project about post-apartheid South Africa, using satellite imagery and computer vision to study the impact of desegregation.
Another DAIR fellow, Milagros Miceli, is researching power imbalances in the data labour industry, focusing on exploitative practices within the data annotation sector. Miceli's work sheds light on the meager wages paid to workers, often located in the Global South, who analyse and label datasets for AI algorithms.
The story of Alex Hanna, another prominent AI ethics researcher, and the emergence of DAIR serves as a call to action for everyone involved in AI. Hanna's resignation letter from Google addressed the "whiteness problem" within tech, arguing that Google's corporate culture was inherently biased and resistant to change. Hanna's vision for a more inclusive and ethical AI landscape is tied to the broader implications of DAIR's work for the future of AI development.
Supporting organisations like DAIR is essential for creating a more equitable and just AI future for all. As Gebru puts it, DAIR seeks to "bring AI back down to earth" — combating the myth of AI as a mysterious, inevitable force and instead framing it as a human-created technology that should be shaped consciously to promote fairness, equity, and humanity.
For more information about DAIR, visit their website at https://www.dainow.org/. The Algorithmic Justice League (https://www.ajlunited.org/) and the Partnership on AI (https://www.partnershiponai.org/) are other organisations working towards ethical AI and addressing the harms caused by AI technologies.
- Timnit Gebru, the founder of DAIR, envisions a future where AI development is independent from profit-driven systems and inclusive of diverse perspectives, prioritizing equity, transparency, and community involvement.
- DAIR's research projects, such as Raesetje Sefala's work on post-apartheid South Africa and Milagros Miceli's research on power imbalances in the data labor industry, aim to use AI as a tool for social good, addressing real-world problems and prioritizing community needs.
- In light of the ethical challenges and harms embedded in AI technologies, organizations like DAIR, Algorithmic Justice League, and the Partnership on AI are essential for creating a more equitable and just AI future, bringing AI back down to earth to be shaped consciously to promote fairness, equity, and humanity.
