Google fires second AI ethics leader as controversy over search, diversity grows

(Reuters) – Google Alphabet Inc unveiled employee scientist Margaret Mitchell on Friday, they both said, a move that strained company divisions on academic freedom and diversity that had been visible since December’s ouster to AI ethics researcher Timnit Gebru.

Google said in a statement that Mitchell violated the company’s code of conduct and security policies by moving electronic files outside the company. Mitchell, who announced her shooting on Twitter, did not respond to a request for comment.

Google’s ethics in artificial intelligence work has been under scrutiny since the firing of Gebru, a scientist who gained a reputation for displaying bias in facial analytics systems. The implementation prompted thousands of Google employees to complain. She and Mitchell had called for more diversity and inclusion among Google search staff and were concerned that the company was starting to question critical papers about their results.

Gebru said she was fired by Google after questioning an order not to publish a study saying AI that imitation language could harm peripheral numbers. Mitchell, who co-authored the paper, publicly criticized the company for burning Gebru and weakening the credibility of its operations.

The pair had for about two years co-led an ethical AI team, started by Mitchell.

Google AI search director Zoubin Ghahramani and a company lawyer informed Mitchell’s shooting about her shooting Friday in a meeting called without warning, according to someone familiar with the case. The person stated that little explanation was given for the appointment. Google declined to comment.

The company said Mitchell fired on control recommendations with auditors and a review committee. They said her breaches “included the deletion of confidential documents sensitive to the business and private data of other employees”. The investigation began on 19 January.

Google employee Alex Hanna said on Twitter that the company was running a “smear campaign” against Mitchell and Gebru, with whom she worked closely. Google declined to comment on Hanna’s comments.

Google has hired top scientists with promises of search freedom, but the limits are tested as researchers continue to write about the negative effects of technology and offer unobtrusive views on their employer’s results. .

Only in December did Reuters report in December that Google introduced a new “sensitive topics” review last year to ensure that papers on topics such as the oil industry and the company’s content recommendation systems would not lead to legal or regulatory trouble. . Mitchell publicly raised concerns that the policy could lead to censorship.

Google reiterated to researchers in a memo and meeting Friday that they were working to develop a pre-publication review of papers. It also announced new policies Friday to address sensitive events and evaluate officers based on diversity and team inclusion.