After a year of trying to recover from the data breaches and fake news epidemic, the company’s content moderation team came to light with reports of “underpaid, stressful and sometimes traumatic” work load.
The latest report by Reuters has said that one of Facebook’s contractor, Genpact has raised minimum wages for the content moderation teams it runs for Facebook in India. The report added that Facebook has unveiled several initiatives in the past six months to support wellbeing in content moderation teams.
These content moderation teams have to wade through billions of potentially harmful or violent posts on its platforms. Genpact has since more than doubled minimum salaries for new recruits to its Facebook teams to INR 2.5 Lakh ($3,503) a year.
Earlier, Reuters had reported that based on job adverts and employee pay stubs, the starting salary for some graduate content moderators was just INR 1 Lakh ($1,401) a year. A Genpact spokesperson said, “We can confirm that Genpact did recently raise salaries for our team in India that supports Facebook as part of an annual review process that emphasizes paying a competitive market rate.”
It was reported that as of February, Genpact employed more than 1,600 employees in its Facebook teams in Hyderabad. These employees review posts in English as well as a host of Indian languages, Arabic, and some Afghan and Asian tribal dialects, according to Facebook. However, it is unclear if the salary hike has been for new or old employees.
Genpact statement reportedly said that the rise “reinforces our commitment to our employees and helps us attract and retain top talent.” Facebook in May raised minimum wages for US content moderators to $18 to $22 per hour depending on location from $15 nationwide, and last week said it is still studying increases in other countries.
Recently, Facebook was under fire for its content moderation activities. At the time, it was reported that Wipro employees from its Hyderabad office and 260 ‘workers’ have been hired by Facebook, through Wipro, to manually tag or label photos, posts, links shared on timeline, stories and videos into several category items according to the ‘five dimensions’ that Facebook considers key for its AI datasets.
Labelling or data annotation is one of the fastest growing industries as companies race to train AI and machine learning systems. Employees label images, text, logos, and symbols to help computers understand context and contents of the image or text. This is then used to develop consumer-facing AI features such as text OCR or object recognition inside camera apps.
At a time when Facebook is standing up against Indian government to ensure data safety of its users, the company’s third party moderation continues to raise questions.