About a year ago, Google fired their star artificial intelligence (AI) researcher. Now, she’s bounced back with a firm of her own that received $3.7 million in funding from the MacArthur Foundation, Ford Foundation, Kapor Center, Open Society Foundation, and the Rockefeller Foundation.
— Distributed AI Research Institute (@DAIRInstitute) December 2, 2021
The Washington Post tells the story of Timnit Gebru, a prominent artificial intelligence computer scientist who just founded Distributed Artificial Intelligence Research Institute (DAIR). The institute, according to the outlet, “aims to both document harms and develop a vision for AI applications that can have a positive impact on the same groups.”
And there’s no better person to lead that charge than Gebru, who was instrumental in noting that facial recognition software has a bias against people of color. Google, for its part, fired Gebru after she published an academic paper critiquing the tech giant on what’s known as “large language models,” which is used to help conversational search queries.
“I’ve been frustrated for a long time about the incentive structures that we have in place and how none of them seem to be appropriate for the kind of work I want to do,” she said to the Washington Post.
DAIR also has a laundry list of influential and powerful people on its board. Safiya Noble, a recent recipient of the MacArthur Genius grant and author of Algorithms of Oppression, is a member of DAIR’s advisory committee, along with Ciira wa Maina, co-founder of Data Science Africa, who has researched food security, climate change, and conservation, according to the Post.
But, as AfroTech previously reported, Google has been under fire for its practices in the past. Recently, a Black Google employee revealed that he was stopped and questioned by security after being reported as a suspicious person by someone at the company. Even their initiatives in “addressing the diversity gap” have been accused of window-dressing by critics.
She tried to bring things to Google's attention in 2019
According to the Washington Post, Gebru first began telling Google about their problems in 2019. At that time, the Ethiopian native with Eritrean heritage said that the tech giant needed non-English expertise in technology like detecting toxicity in online comments.
Gebru said that her observations were met with pushback from the higher-ups, who said that there was “sufficient expertise” on international audiences.
Google had no comment for the Post.