An initiative by Facebook, Microsoft, Twitter and Google’s YouTube to fight the spread of terrorist content online appears to be gaining momentum.
A database that the four organizations established last December to share information with each other and with other technology platforms now contains more than 40,000 “hashes” that uniquely identify videos and images containing extremist content. YouTube and the other members of the initiative are using the hashes to identify, remove and, in some cases, block videos and images that violate each company’s respective policies with regard to terror content.
The companies are sharing the hashes as part of a Global Internet Forum to Counter Terrorism (GIFCT) initiative that they announced at the EU Internet Forum last December and formally launched in June 2017. Any technology company can participate in the effort and add their own hashes of terrorist images or videos to the database. Participating companies can also use hashes in the shared database to identify extremist content on their own services.
Since launching the database, multiple other internet companies have joined the hash-sharing consortium, Google said in an update on its public policy blog Dec. 4. Among them are LinkedIn, Instagram, Cloudinary, Snap, Oath, sk.fm and Justpaste.it. Google and the other founding members of GIFCT are currently working on getting more platforms to participate in the effort, the company noted.
“In order to disrupt the distribution of terrorist content across the Internet, companies have invested in collaborating and sharing expertise with one another,” the Google update said. “GIFCT’s knowledge-sharing work has grown quickly in large measure because companies recognize that in countering terrorism online we face many of the same challenges.”
As part of GIFCT, Google and the other founding members had hoped to work with about 50 companies in 2017 and share best practices with them on how to prevent the dissemination of extremist content on their platforms. The GIFCT group has exceeded that goal, having worked with 68 companies so far this year.
In addition to the shared database, GIFCT members are also exploring other new technological approaches and knowledge-sharing initiatives to combat the spread of terror propaganda online. As part of that effort, the group hosted a meeting in August where representatives from government organizations, nongovernmental organizations and technology companies convened to discuss some of these new approaches.
The GIFCT initiative stems from mounting concerns over the use of YouTube, Facebook, Twitter and other platforms by terror groups to disseminate extremist content and to find new recruits for their causes. Google and other tech giants have committed to using their fullest resources to identify, remove and block terror content on their platforms.
In recent months, Google for instance has stepped up the use of machine learning tools to identify potentially troublesome content on YouTube and its other platforms.
The company is working with numerous NGOs and other entities to manually review content that gets flagged for inspection by the automated machine learning tools. Google has also started making it harder for users to find videos and images that are on the borderline between acceptable and extremist content and for the uploaders of these videos to profit from them via ads.