After receiving strong criticism from leaders of European countries, as well as from its own community, Facebook is preparing artificial intelligence to combat messaging by terrorists and generally limit all propaganda on its platform.
AI technology will not only be applied to identify Muslim extremists like ISIS, but any group that promotes violent behavior or engages in acts of violence. A broader definition of terrorism could include violent gang activities, drug trafficking, or white nationalists who advocate violence.
Facebook is currently testing the use of natural language processing and analysis by adding features that identify violent posts based on the words used by accounts that have already been suspended.
“Today we are experimenting with analyzing text that we have already removed to identify terrorist organizations like ISIS or Al Qaeda. We are trying to identify messages based on the text using an algorithm that is still in the early stages of learning,” said Facebook’s director of global policy management Monika Bickert and director of counterterrorism policy Brian Fishman in a blog post.
The social network highlighted a number of initiatives it has already taken to control extremist content, such as using technology to identify photos and videos associated with terrorists, deleting fake accounts from Facebook and seeking advice from a group of experts on combating terrorism.
Earlier this month, following the terrorist attack in London that killed seven people, British Prime Minister Theresa May called on nations to create international agreements to “regulate cyberspace to prevent the spread of extremist and terrorist planning,” calling on Facebook to take immediate action. Theresa May blamed the social network in part for the crime because its platforms give terrorists a “safe place” that helps them operate.
In March, a German minister said that companies like Facebook and Google could face fines of $53 million if they fail to curb hate speech.
Shortly after the events of May, Facebook's director of policy, Simon Milner, said the company would make its social network a "hostile environment" for terrorists.
Last month, Mark Zuckerberg said his company plans to hire 3,000 more people for its Facebook Community Operations, up from 4,500 it currently employs. The team reviews content that is deemed potentially inappropriate or violates Facebook policy.
In its latest blog post, Facebook is asking for help from anyone who thinks they can help. So if you have ideas on how to stop the spread of terrorism online, you can write to hardquestions@fb.com.
