SAN FRANCISCO, Dec 5, 2020 (BSS/AFP) – More than 1,400 Google employees on
Friday demanded that the tech giant explain why it dismissed a Black
artificial intelligence ethics researcher.
Timnit Gebru tweeted early in the week that Google told her a resignation
she had not proffered was accepted, after she sent a message to an internal
group lamenting the “silencing of marginalized voices.”
She had also told the “Brain Women and Allies” group in the email that she
had been ordered to retract a research paper.
Gebru is an outspoken diversity advocate and co-founder of a group devoted
to increasing Black talent in artificial intelligence.
Until Tuesday, she was a staff researcher and co-lead of an ethical AI
team at Google. Gebru tweeted what she said was the dismissal message from
Google, which noted that “aspects of the email you sent last night to non-
management employees in the brain group reflect behavior that is inconsistent
with the expectations of a Google manager.”
The research paper Gebru was told to retract looked at potential for an AI
tool used by Google and other tech firms to imitate human writing of hate
speech and biased language, according to an NPR report.
Slightly more than 1,400 Google employees were among the nearly 3,300
names on an online letter calling on the tech giant to explain Gebru’s
dismissal along with the reason for ordering her to withdraw her research.
The letter demanded Google make an “unequivocal” commitment to research
integrity and academic freedom.
“Instead of being embraced by Google as an exceptionally talented and
prolific contributor, Dr Gebru has faced defensiveness, racism, gaslighting,
research censorship, and now a retaliatory firing,” the letter contended.
– ‘Difficult moment’ –
In response to a request for comment, Google referred AFP to an email by
AI division leader Jeff Dean to researchers at the company that he posted
publicly on Friday.
Dean said in the email it was a “difficult moment” and he wanted to clear
up what he saw as “a lot of speculation and misunderstanding on social media”
about Gebru’s departure from the company.
The research paper authored by Gebru and other researchers was shared for
review just a day before a submission deadline and “had some important gaps
that prevented us from being comfortable putting Google affiliation on it,”
Dean said in the email, a copy of which was posted online.
Gebru responded with conditions — including that Google give her the
names of everyone involved in reviewing the paper — saying she would leave
if they were not met, Dean contended.
“We accept and respect her decision to resign from Google,” Dean said in
the email.
“Given Timnit’s role as a respected researcher and a manager in our
Ethical AI team, I feel badly that Timnit has gotten to a place where she
feels this way about the work we’re doing.”
Gebru, who is Ethiopian American, earned engineering degrees and a
doctorate from Stanford University in Silicon Valley. She has worked at Apple
and conducted research at Microsoft.
Gebru’s research has included showing the likelihood of facial detection
technology making mistakes when identifying people of color.
Her dismissal came just days after a US labor board complaint accused
Google of using surveillance, interrogation and other tactics to spy on
activist employees.
The National Labor Relations Board (NLRB) complaint stemmed from the
dismissal a year ago of a quartet of employees dubbed the “Thanksgiving
Four.”
The workers sought a federal investigation into their dismissal, alleging
they were sacked in retaliation for their labor organizing efforts, while
Google maintained the employees had violated data security policies.