Meta received a total of 55,497 requests for user data from the Indian government in the first half of Calendar 2022. These figures are more than the approximately 50,382 inquiries the social media company received between July and December 2021.
During the first six months of 2022, requests for user data from governments worldwide increased by 10.5 percent, from 214,777 to 237,414, according to Meta. The US continues to send the most requests, followed by India, Germany, Brazil, France and the UK.
Meta added that it has restricted access to 23 articles reported by the Election Commission of India. The restrictions came after election complaints were filed under the Indian Penal Code, the Representation of the People Act, about issues such as the spread of disinformation by candidates and incitement to sectarian violence.
19 of them are limited only temporarily (during the designated lighting period).
“We have also restricted access to 597 items in response to Ministry of Electronics and Information Technology (MeitY) directives under Section 69A of the Information Technology Act, 2000 for violations including content against public security and public order. (We have also restricted access to six items in response to instructions from the Ministry of Information and Broadcasting for violating Regulation 16 of the Information Technology (Mediation Guidelines and Digital Media Code of Ethics) Rules, 2021,” said the Meta Integrity and Transparency Report, 3-3-2022 quarter.
Meta also restricted access to 71 articles due to other court orders, 13 articles for IP infringement, and two articles in response to personal defamation reports.
As part of compliance with the new IT Rules for September 2021, Meta said it has received 587 reports through the Indian complaint mechanism for Facebook and responded to 100 percent of them.
In the case of Instagram, Meta India received 1115 reports through the complaint mechanism and responded to all of them. The most complaints were in the category of fake profiles (518), hacked accounts (332).
Globally, the report said the prevalence of content related to violence and harassment remained low, at 0.08 percent on Facebook and between 0.04 and 0.05 percent on Instagram.
Actions against hate speech content on Facebook fell from 13.5 million to 10.6 million in Q3 2022, the company added. “We did this by using data from past user requests to identify posts that might have been mistakenly deleted without the appropriate cultural context. For example, we can now better recognize humorous terms of endearment used between friends, or better detect words that may be offensive or inappropriate in one context but not in another. As we improved this accuracy, our rate of proactive detection of hate speech also decreased from 95.6 percent to 90.2 percent in Q3 2022,” the company said.