Because many Google employees are in revolt
Storm of criticism on Google after the exit from the company of Timnit Gebru, a pioneer in the study of ethics in artificial intelligence
Employees and academics in revolt against Big G.
Over 1,200 Google employees and more than 1,500 academic researchers have signed an open letter of protest against the Mountain View giant.
The reason is the exit from the company of Timnit Gebru, one of the best known and most respected African American scientists in the field of artificial intelligence ethics, who denounced on Twitter that she was fired by Google for expressing criticism of her efforts on diversity. and highlighted the biases of AI systems.
Google told employees that she resigned.
Timnit Gebru helped improve Google's public image as a company that elevates black computer scientists and questions the harmful uses of artificial intelligence.
All the details.
WHAT HAPPENED BETWEEN GOOGLE AND TIMNIT GEBRU
Timnit Gebru was technical co-head of Google's Ethical AI team until last Wednesday.
Gebru wrote on the social network that she was torpedoed after sending an email (titled “Silence the marginalized voices in any way possible”) to an internal women's group and to other colleagues working in the company's AI unit.
An email expressing frustration about Google's diversity programs and evoking a 2018 study of her on racial and gender bias in facial recognition software, which a company executive asked her to retract last November.
THE DOCUMENT OF DISCORD
Gebru had recently worked on a paper examining the risks of developing computer systems that scan huge human language databases. The study mentions Google's new technology used in its research activity.
In addition to pointing out the potential dangers of racial prejudice, the paper also cited the environmental cost of wasting so much energy running the models. An important issue for a company that prides itself on its commitment to being carbon neutral since 2007 while striving to go even greener.
LEAVING THE COMPANY LEADED BY SUNDAR PICHAI
The scientist reported that she attempted to negotiate with Google, offering to remove her name from the study in exchange for a full explanation of the company's objections and a discussion of how to better manage these issues in the future, threatening to resign if she fails. reply.
Google denied her request and informed her that it accepted her resignation.
“I felt like I was being censored and thought this could have implications in all ethical research on AI,” Gebru told Wire.
Apparently my manager's manager sent an email my direct reports saying she accepted my resignation. I hadn't resigned — I had asked for simple conditions first and said I would respond when I'm back from vacation. But I guess she decided for me
that's the lawyer speak.
– Timnit Gebru (@timnitGebru) December 3, 2020
THE POSITION OF JEFF DEAN, HEAD OF GOOGLE AI
As CNN reports, in an email sent Thursday to Google Research employees it posted on Friday, Jeff Dean, head of Google's artificial intelligence, told employees his point. “Gebru co-authored an article but didn't give the company the two weeks to review it before its deadline. The document was revised internally – he wrote – but "did not meet our requirements for publication".
Dean added: “Unfortunately, this particular document was only shared with one day's notice before the deadline. We need two weeks for this type of review and then instead of waiting for reviewers' feedback, it has been approved for submission and sent. ”
Hence the "accepted" resignation for Gebru.
THE WORK OF GEBRU IN THE FIELD OF AI
Timnit Gebru is famous in the world of artificial intelligence ethics. He also worked at Apple developing the algorithms for the first iPads and earned his doctorate from Stanford Artificial Intelligence Laboratory.
She is co-founder of the Black in AI group, which promotes black employment and leadership in the industry. She is known for the 2018 study co-signed with Joy Buolamwini, a researcher at the Massachusetts Institute of Technology, which found racial and gender bias in facial recognition software .
WHAT END OF THE "DON'T BE EVIL" MOTTO?
The story risks being a bad blow to the image and credibility of the company led by Sundar Pichia, whose original motto was “Don't Be Evil” (“Don't hurt”).
THE PREVIOUS PROTESTS OF GOOGLE EMPLOYEES
Gebru's abrupt departure adds to years of tensions, including several resignations and layoffs, in the AI department and other Google organizations due to diversity and if the company's efforts to minimize the potential harm of its services are sufficient.
In the summer of 2018, Google abandoned the Pentagon's Maven project. The project uses computer vision software to automatically analyze footage collected by US military drones. Big G was forced into the decision by employee protests claiming Big G is out of the war business.
Also in 2018, Google employees signed an open letter inviting the company to cancel the Dragonfly project regarding a search engine banned for the Chinese market that would have allowed state surveillance.
THE NEXT CHALLENGES OF BIG G
How it manages its initiative on AI ethics and the internal dissent sparked by Gebru's exit will be the first hurdle the Mountain View giant will face in the new year. Without forgetting the accusation by the US Justice Department of taking advantage of its dominant position to stifle competition.
This is a machine translation from Italian language of a post published on Start Magazine at the URL https://www.startmag.it/innovazione/perche-molti-dipendenti-di-google-sono-in-rivolta/ on Tue, 08 Dec 2020 07:43:20 +0000.