Meta has received a total of 55,497 requests by the government of India on user data during the first half of Calendar 2022. These numbers are up from around 50,382 requests the social media company received in the July-December 2021 period.
According to Meta, during the first six months of 2022, requests from governments across the globe for user data were up 10.5 per cent from 214,777 to 237,414. The US continues to submit the largest number of requests, followed by India, Germany, Brazil, France and the UK.
Meta added that it restricted access in India to 23 items reported by the Election Commission. The restrictions came into force after electoral complaints on issues such as spreading candidate misinformation and promoting sectarian violence were made under the Indian Penal Code, Representation of the People Act.
Of these, 19 items were restricted only temporarily (during the designated blackout period).
“We also restricted access to 597 items in response to directions from the Ministry of Electronics and Information Technology (MeitY) for violations under Section 69A of the Information Technology Act, 2000, including content against security of the state and public order. (We also restricted access), to six items in response to directions from the Ministry of Information & Broadcasting for violating Rule 16 of the Information Technology (Intermediary Guidelines & Digital Media Ethics Code) Rules, 2021,” said Meta in its Integrity and Transparency Report, Q3 2022.
Meta also restricted access to 71 items due to other court orders, to 13 items for IP infringement, and to two items in response to private reports of defamation.
As part of the compliance under the new IT Rules, 2021, for the month of September, Meta said for Facebook it received 587 reports through the Indian grievance mechanism, and that it responded to 100 per cent of these.
In the case of Instagram, Meta received 1,115 reports through the Indian grievance mechanism, and responded to all of them. The highest complaints were in the category of fake profiles (518), and followed by accounts hacked (332).
Globally, the report said that the prevalence of bullying and harassment-related content remained low, at 0.08 per cent on Facebook and between 0.04-0.05 per cent on Instagram.
The company added that its actions against hate speech content dipped from 13.5 million to 10.6 million in Q3 2022 on Facebook due to improvement in the accuracy of its AI technology. “We’ve done this by leveraging data from past user appeals to identify posts that could have been removed by mistake without appropriate cultural context. For example, now we can better recognise humorous terms of endearment used between friends, or better detect words that may be considered offensive or inappropriate in one context but not another. As we improved this accuracy, our proactive detection rate for hate speech also decreased from 95.6 per cent to 90.2 per cent in Q3 2022,” said the company.