LinkedIn received 10 requests from Indian government for user data in 2019, but did not provide data in response to those requests, the social media network for professionals disclosed in its transparency report. Those ten requests requested LinkedIn for data on 21 users, and LinkedIn did not provide any of the data requested. LinkedIn says it has not provided any data in response to government requests from India since 2011, the earliest year for which such data is available.
India was one of eight countries to send LinkedIn a content takedown request in the second half of 2019, sending one such request. LinkedIn complied with that request, though the company did not disclose whether it was a judicial order or a government request. In 2018, the company received three takedown requests from India, and complied with two.
In the US, LinkedIn disclosed that they provided data on around 1,458 user accounts in 2019; the company said it received 0–499 national security letters (which are only disclosable in multiples of 500), and that they would start disclosing orders under the US Foreign Intelligence Surveillance Act for June–December 2019 six months later, as the law requires such a delay.
32.8 million fake accounts thwarted in 2019
Aside from government requests, LinkedIn has started providing community action it has been taking, starting from 2019, in six month intervals.
- Most fakes stopped by automated filters: LinkedIn said it removed or prevented the creation of 32.8 million fake accounts in 2019, the vast majority (27.3 million) of which it detected automatically before registration. 5.2 million were “restricted proactively” before being reported by users, and 153,000 were removed based on user reports. In the second half of 2019, LinkedIn said, 93% of fake accounts were prevented from being created or removed automatically, while 7% were removed after manual review.
- 99% of spam stopped filtered automatically: In 2019, LinkedIn automatically detected and blocked 143 million instances of “spam and scams”. Apart from this, around 212,500 instances were removed after manual reporting and review.
- Content violations: During the year, the company removed 32,261 instances of harassment; 20,356 instances of adult content; 3,772 pieces of violent/graphic content, 2,137 instances of hate speech, and 189 instances of child exploitation.
- Copyright takedowns: Throughout 2019, the company accepted 99.98% of copyright infringement reports it received, rejecting only 62 such reports. It received 20,409 requests throughout the year reporting 406,271 infringements.
Read the transparency report here.