Google researcher says she is in A.I.


A well-respected Google researcher said she was fired from the company after criticizing the approach to minority attitudes and the prejudices built into today's artificial intelligence systems.

Timnit Gebru, Co-Head of Google's Ethical A.I. Team said in a tweet on Wednesday night that she was fired over an email she sent the day before to a group that included company employees.

In the email verified by the New York Times, she expressed anger at Google's response to efforts by her and other employees to raise minority attitudes and raise awareness of prejudice about artificial intelligence.

“Your life just keeps getting worse when you stand up for underrepresented people. You are starting to upset the other executives, ”read the email. "There's no way more documents or conversations could make a difference."

Her departure from Google underscores the growing tension between outspoken Google employees and buttoned management, and expresses concerns about the company's efforts to build fair and reliable technology. It can also have a chilling effect on both black technicians and researchers who have left science in recent years to get high-paying jobs in Silicon Valley.

"Your dismissal only shows that scientists, activists, and scientists who want to work in this field – who are black women – are not welcome in Silicon Valley," said Mutale Nkonde, a Stanford Digital Civil Society Lab employee. "It's very disappointing."

A Google spokesman declined to comment. In an email to a Google employee, Jeff Dean, who runs the A.I. The work, including that of Dr. Gebru and her team, described her departure as “a difficult moment, especially given the important research topics she was involved in and the extent to which responsible A.I. Research as an organization and as a company. "

After years of having employees holding free-wheeling discussions in company-wide meetings and online message boards, Google has begun cracking down on discourse in the workplace. Many Google employees have resisted the new restrictions, arguing that the company has deviated from a tradition of transparency and free debate.

On Wednesday, the National Labor Relations Board said Google had most likely breached labor law when it fired two employees who were involved in organizing work. The federal agency said Google illegally monitored employees before they were fired.

Google's struggles with its employees, who have spoken in recent years about the company's handling of sexual harassment and its collaboration with the Defense Ministry and federal border agencies, have diminished its reputation as a utopia for technicians with generous salaries, perks and freedom of work.

Like other tech companies, Google has been criticized for not doing enough to address the shortage of women and ethnic minorities among its ranks.

Racial inequality issues, particularly the mistreatment of black workers in tech companies, plagued Silicon Valley for years. Coinbase, the most valuable cryptocurrency start-up, has seen an exodus of black workers over the past two years because workers said it was racist and discriminatory treatment.

Researchers fear that the people who build artificial intelligence systems could build their own prejudices into the technology. In recent years, several public experiments have shown that the systems often interact differently with people of color – possibly because they are underrepresented among the developers who create these systems.

Dr. Gebru, 37, was born and raised in Ethiopia. As a researcher at Stanford University in 2018, she helped write what is widely regarded as a turning point in efforts to identify and remove biases in artificial intelligence. She joined Google later that year and helped build the Ethical A.I. Team.

After hiring researchers like Dr. Gebru has chosen Google as a company for "ethical" A.I. However, it is often reluctant to publicly acknowledge deficiencies in its own systems.

In an interview with The Times, Dr. Gebru said her outrage was due to the treatment of a research paper she wrote with six other researchers, four of them at Google. The article, also discussed by The Times, highlighted shortcomings in a new generation of language technology, including a system developed by Google to support the company's search engine.

These systems learn the vagaries of language by analyzing enormous amounts of text, including thousands of books, Wikipedia entries, and other online documents. Because this text contains biased and sometimes hateful language, the technology can generate biased and hateful language.

After she and the other researchers presented the paper to an academic conference, Dr. Asked a Google manager to either withdraw the paper from the conference or remove her name and the names of the other Google employees. She refused to do so without further discussion, saying in the email sent Tuesday evening that if the company couldn't explain why they were withdrawing the paper and address other concerns, she would step down after a reasonable time should.

The company replied to her email that it could not meet her demands and that her resignation would be accepted immediately. Your access to corporate email and other services was immediately revoked.

In his statement to employees, Dean said that Google had "respected their decision to resign". Mr Dean also said the paper does not recognize the latest research showing ways to reduce bias in such systems.

"It was dehumanizing," said Dr. Gebru. “You may have reasons to stop our research. Most annoyingly, however, they refuse to discuss why. "

Dr. Gebru is leaving Google at a time when A.I. Technology plays a bigger role in almost every area of ​​Google's business. The company has focused its future on artificial intelligence – be it with its voice-activated digital assistant or its automated placement of advertisements for marketers – as a breakthrough technology to make the next generation of services and devices smarter and more powerful.

Sundar Pichai, CEO of Alphabet, the parent company of Google, has compared the emergence of artificial intelligence to that of electricity or fire and said it is essential to the future of the company and the computer. Earlier this year, Mr Pichai called for stronger regulation and responsible use of artificial intelligence, arguing that society needs to balance potential harm with new opportunities.

Google has made repeated commitments to removing bias in its systems. The problem, said Dr. The problem is that most of the people who make the final decisions are men. "Not only are they failing to recruit more people from minority communities, they are suppressing their voices," she said.

Julien Cornebise, Honorary Associate Professor at University College London and former researcher at DeepMind, a prominent A.I. The lab, run by the same parent company as Google's, was among many artificial intelligence researchers who said Dr. Gebru & # 39; s departure reflected a bigger problem in the industry.

"This shows how some big tech companies only support ethics and fairness and other good causes as long as their positive PR effects outweigh the extra scrutiny they bring," he said. “Timnit is a brilliant researcher. We need more like them in our field. "