The events unfolded as Timnit took it to Twitter to tell everyone about the incidences that led to her layoff. Gebru was sacked because she refused to take her name off a paper that raised some ethical concerns regarding the potential risks associated with large language processing models.
Sign up for your weekly dose of what's up in emerging technology.
Apart from working in the ethical AI team, Timnit Gebru was also very vocal about Google’s lack of diversity and inclusion of minority communities, which according to her might have been one of the reasons she was fired.
Her layoff has led to massive outrage in the industry, generating a discussion of Google’s priorities in ethical AI and the discrimination faced by women and minorities in the tech community.
Is Big Tech Really Concerned About Ethical AI
The ethical AI community has always been very critical of big tech firms. Deployment of biased algorithms in the past, priority of profits over ethics, and lack of effort towards inclusivity and diversity by these firms have made them sceptical.
Cathy O’Neal, the founder of O’Neil Risk Consulting & Algorithmic Auditing, said that she always gave the benefit of the doubt to Google simply because of Gebru and her team’s work. However, now Google no longer deserves that, said O’Neal.
Ellen K. Pao, former CEO of Reddit and the founder of Project Include, also expressed solidarity with Gebru. She highlighted the importance of protecting employees that raise ethical concerns which can prevent companies from developing products that might harm marginalised communities.
Gebru’s immediate manager, Samy Bengio, as well posted on Facebook, expressing his shock over the layoff. He showed strong support of her scientific work in AI while also expressing solidarity with her and the ethical AI team.
Several other people in the tech and ethical AI, including Joy Buolamwini, the founder of Algorithm Justice League and co-author of the paper that was rejected by Google, came forward to express their support for Gebru.
The ‘Google Walkout for Real Change’, an employee activism group, has also filed a petition, signed by more than 7000 people, which demands more transparency and an unequivocal commitment by Google to research integrity and academic freedom.
Despite the consequences, Timnit Gebru’s layoff has forced people to stick out their necks and question the tech companies’ seriousness over ethical AI.
Employee Discrimination Reflecting In Algorithms
Over the years, Gebru and other AI ethicists have shown how discrimination in workplaces and lack of inclusivity reflects in algorithms. Her rejected paper also showed results on over-reliance on data of the richest countries and how that will reflect practices only of those nations.
To this, the employees of Google and other tech companies have voiced out their opinions on how Gebru tried her best to create a safe environment. At the same time, they expressed how the lack of safe and inclusive workplaces in tech companies is leading to a lack of diversity in AI teams.
Of the many who voiced out their opinions, Negar Rostamzadeh stated how Timnit was able to create a safe zone for underrepresented researchers and how she helped in educating people in the organisation.
Many also expressed their concern over the existing discrimination against women and people of colour, and how the lack of it is playing a significant role in creating unethical algorithms.
Alongside current and former employees of Google also spoke about how the company uses its HR and PR mechanisms to suppress the voice of employees who try to stand against ethical concerns.
Leslie Miley, a former Google employee who now works as the CTO at Obama Foundation, explained how Google fires its employees who raise their voices on critical issues. Miley told that high ranking individuals usually tend to undermine such employees in front of senior leadership, including HR, which basically ‘gives them a reason’ to fire them.
Gebru’s teammate, Margaret Mitchell, also tweeted about the company’s double standards, as the team is usually efficient at making them look good but could not use Gebru’s research work to do the same.
On Tuesday, ‘Google Walkout for Real Change’ expressed solidarity by criticising Jeff Dean, the person responsible for her layoff. As a matter of fact, the industry believes that concerns over discrimination against women and people of colour and use of mechanisms to fire employees who raise ethical issues do not portray a genuine intention of the company to produce ethical AI.
While Google did not want the paper to come in the public domain, the sacking of Timnit Gebru did the exact opposite. It has started a very significant discussion in the tech community about the seriousness of the big tech to produce ethical AI, protection of people who want to raise concerns around the topic and ensuring inclusivity to avoid algorithmic biases.
These industry reactions on Timnit Gebru’s layoff, thus, highlights a significant phase in the tech field. This might help tech firms realise the criticality of producing ethical algorithms and creating a safe and inclusive environment for employees.
With the outrage that followed Gebru’s sacking, the Google CEO, Sundar Pichai apologised on Wednesday for her abrupt exit. He said that Google ‘needs to accept responsibility’ over the fact that a prominent Black female leader with immense talent left Google unhappily.