The FBI has signed a contract with Dataminr, a company that sifts through the Twitter newsfeed to identify issues and trends in real time. Dataminr will provide the bureau with an advanced alerting tool with licenses for over 200 users at the agency that will give the FBI the capability to search the complete Twitter firehose in real time, using customized filters.
Dataminr also agreed to provide 24-hour support in creating searches and filters, and helping the agency to troubleshoot issues.
Only an estimated 1 percent of tweets are made available to the public through Twitter’s API. Using Dataminr, the FBI will now have access to all messages written on Twitter, and will be able to search through the messages for key words and phrases, providing a level of social media monitoring that had not been unavailable until now. They will be able to use Dataminr’s proprietary algorithms and geolocation tools to sift through the entirety of messages posted by Twitter’s 317 million average monthly users worldwide.
In their initial request for this service, the FBI noted that social media platforms are used by subjects of FBI investigations, and that the information sent by “terrorist groups, intelligence services and criminal organizations to further their illegal activity creates a demonstrated need for tools to properly identify the activity and react appropriately.” Criminal activities could include communication, fundraising and recruitment activities by these groups which are conducted through social media.
In May of this year, Twitter blocked Dataminr from entering into a similar agreement with the CIA. Dataminr, which is 5% owned by Twitter, had also accepted investment from In-Q-Tel, the non-profit investment arm of the CIA, and were conducting an unpaid system test for the agency. However, Dataminr ended the contract with the CIA due to a long-standing policy at Twitter barring the government from using the social media platform as a surveillance tool.
The policy is part of Twitter’s Developer Agreement, which states specifically that surveillance of Twitter users is prohibited in situations that would normally require a court order or subpoena. It also specifically bans any use of Twitter API or Twitter content by a government entity.
Twitter has taken its own steps to combat violent extremism, by suspending over 360,000 from 2015-2016 for violating the company’s policy against the promotion of terrorism. Twitter also noted that they have improved their response time in deactivating problem accounts and have made progress in preventing those with suspended accounts from immediately returning under a different name. The company also reiterated its commitment to work with law enforcement agencies under the appropriate circumstances, using valid legal processes including subpoenas or court orders.