In the wake of last week’s revelations that Facebook allowed researchers with direct links to the Department of Defense to conduct manipulative experiments on its users by influencing their news feeds, new reports have emerged detailing how the Pentagon is funding dozens more studies into how to influence and control social media.
Reporters with The London Guardian reveal that DARPA, the Pentagon-run Defense Advanced Research Projects Agency, is engaged in multiple programs targeting Facebook, Twitter, Pinterest, Kickstarter and other popular social media sites.
The source is DARPA’s own website, where the shadowy research arm of the military has posted a list of projects funded under its Social Media in Strategic Communication (SMISC) program, with links to documents.
“The general goal of the Social Media in Strategic Communication (SMISC) program is to develop a new science of social networks built on an emerging technology base,” the website states.
“Through the program, DARPA seeks to develop tools to support the efforts of human operators to counter misinformation or deception campaigns with truthful information.”
Of course, what the Pentagon considers to be “truthful information” can just as easily be described as outright propaganda.
The studies, conducted by researchers at institutes such as the University of Southern California, Georgia Tech, and even within companies such as IBM, were all funded with military money.
The Guardian report notes:
While some elements of the multi-million dollar project might raise a wry smile – research has included analysis of the tweets of celebrities such as Lady Gaga and Justin Bieber, in an attempt to understand influence on Twitter – others have resulted in the buildup of massive datasets of tweets and additional types social media posts.
Within the project list is a research aimed squarely at documenting how The Occupy protest movement used Twitter for organisation, as well as research on tracking internet memes, and analysis of how so called “influence behavior” (liking, following, and retweeting) spreads information on several popular social media platforms including Pinterest, Twitter, Kickstarter, Digg and Reddit.
The report notes that DARPA spent around $19 million to fund the research, with $8.9 passing through IBM and distributed to various researchers, and a further $9.6 million going through educational institutions.
Some of the research funneled through IBM culminated in a study titled “Modeling User Attitude toward Controversial Topics in Online Social Media”. This clearly indicates that the military is interested in having the capability to influence opinion via social media.
Within that study, posted on DARPA’s website, the authors describe how their research could allow the government to use social media to encourage and mould opinions.
“For example, a government campaign on Twitter supporting vaccination can engage with followers who are more likely to take certain action (eg spreading a campaign message) based on their opinions.” the study states.
In an even more disturbing example the authors note “…when anti-government messages are spread in social media, government would want to spread counter messages to balance that effort and hence identify people who are more likely to spread such counter messages based on their opinions.”
Of course, DARPA claims there is no wrong in conducting such research. In a statement to The Guardian a DAPRA spokesperson wrote “Social media is changing the way people inform themselves, share ideas, and organize themselves into interest groups, including some that aim to harm the United States… DARPA supports academic research that seeks to understand some of these dynamics through analyses of publicly available discussions conducted on social media platforms.”
When asked why the military should be involved in funding such research, the DARPA spokesperson said that the studies are essential to US defense interests and that some people engaging in the use of social media platforms “aim to to harm the United States.”
The US government in conjunction with the military has long been concerned with studying the potential for civil unrest and how it can control, facilitate and combat it. The U.S. Army War College, in conjunction with numerous think tanks have somewhat obsessively studied the subject for years.
In recent years, the rise of social media, and its potential use for growing and organising protest movements, has spurred a new urgency within government and the military to adapt to and co-opt such tools.
The so-called “spring” revolutions of recent years have been heavily centered around the use of social media, with many even suggesting that Western government and military forces have, at least in part, controlled and even initiated unrest in other parts of the world for strategic purposes by employing social media.
Reports have recently emerged of the Military setting up fake Twitter-like social networks in countries such as Cuba, in order to manipulate and sway popular opinion there. The Military also developed so called ‘sock puppet’ software to create fake online identities and spread propaganda at home and abroad.
The government has also heavily invested in companies that monitor social media and track how opinions and information spreads on such networks.
Facebook was and is deeply connected to the NSA’s PRISM program. Via leaked information, and by the NSA’s own admission, it has been noted that Facebook not only knew about, but also cooperated with the mass spying program. The NSA even masqueraded as Facebook via fake servers, using them as a launching pad to grab information from hard drives, in order to infect millions of computers around the world with malware as part of the mass surveillance program.
The trend is clear – the government and the military have set about fully co-opting social media and turning it into a tool for social control and manipulation.
You don’t need to have 5,000 friends of Facebook to know that social media can have a notorious mix of rumor, gossip and just plain disinformation. The Pentagon is looking to build a tool to sniff out social media propaganda campaigns and spit some counter-spin right back at it.
On Thursday, Defense Department extreme technology arm Darpa unveiled its Social Media in Strategic Communication (SMISC) program. It’s an attempt to get better at both detecting and conducting propaganda campaigns on social media. SMISC has two goals. First, the program needs to help the military better understand what’s going on in social media in real time — particularly in areas where troops are deployed. Second, Darpa wants SMISC to help the military play the social media propaganda game itself.
This is more than just checking the trending topics on Twitter. The Defense Department wants to deeply grok social media dynamics. So SMISC algorithms will be aimed at discovering and tracking the “formation, development and spread of ideas and concepts (memes)” on social media, according to Darpa’s announcement.
Not all memes, of course. Darpa’s not looking to track the latest twists on foul bachelor frog or see if the Taliban is making propaganda versions of courage wolf. Instead, it wants to see what ideas are bubbling up in among social media users in a particular area — say, where American troops are deployed.
More specifically, SMISC needs to be able to seek out “persuasion campaign structures and influence operations” developing across the social sphere. SMISC is supposed to quickly flag rumors and emerging themes on social media, figure out who’s behind it and what. Moreover, Darpa wants SMISC to be able to actually figure out whether this is a random product of the hivemind or a propaganda operation by an adversary nation or group.
Of course, SMISC won’t be content to just to hang back and monitor social media trends in strategic locations. It’s about building a better spin machine for Uncle Sam, too. Once SMISC’s latches on to an influence operation being launched, it’s supposed to help out in “countermessaging.”
Darpa’s announcement talks about using SMISC “the environment in which [the military] operates” and where it “conducts operations.” That strongly implies it’s intended for use in sensing and messaging to foreign social media. It better, lest it run afoul of the law. The Smith-Mundt Act makes pointing propaganda campaigns at domestic audiences illegal.
What exactly SMISC will look like it its final form is hard to say. At the moment, Darpa is only in the very beginning stages of researching its social media tool. They’re focused on researching the brains of the program — the algorithms and software that’ll identify, locate and make sense of social media trends.
For that, they need some social media data to play around with and test on. Darpa wants bidders to create it in one of two ways. Bidders can round up a few thousand test subjects willing to let their social media data be a guinea pig for SMISC’s software. Alternatively, they can rope in some consenting test subjects for a massively multiplayer role playing game in which generating social media data is a key part of gameplay.
SMISC is yet another example of how the military is becoming very interested in what’s going on in the social media sphere. Darpa has plans to integrate social media data into its manhunt master controller, Insight. NATO has already been paying keen attention to Twitter, using data from the micro-blogging service as an intel source to aid in bomb targeting decisions.
Darpa’s presolicitation offers a very vaguely-sourced anecdote spelling out how SMISC could be used. It details how a social media rumor about the location of a particularly reviled individual — identity and location undisclosed — almost led a lynch mob to storm a house in search of him. Authorities who happened to be paying attention to the Internet rumor were fortunate enough to spot it in time to intervene. In this telling of SMISC’s potential applications, the software could be used to as a tripwire to stop potentially dangerous social media campaigns in their tracks.
But we’re sure you — and the Pentagon — can think of a lot less anodyne uses for Darpa’s social media propaganda tool.
The US military is developing software that will let it secretly manipulate social media sites by using fake online personas to influence internet conversations and spread pro-American propaganda.
A Californian corporation has been awarded a contract with United States Central Command (Centcom), which oversees US armed operations in the Middle East and Central Asia, to develop what is described as an “online persona management service” that will allow one US serviceman or woman to control up to 10 separate identities based all over the world.
The project has been likened by web experts to China’s attempts to control and restrict free speech on the internet. Critics are likely to complain that it will allow the US military to create a false consensus in online conversations, crowd out unwelcome opinions and smother commentaries or reports that do not correspond with its own objectives.
The discovery that the US military is developing false online personalities – known to users of social media as “sock puppets” – could also encourage other governments, private companies and non-government organisations to do the same.
The Centcom contract stipulates that each fake online persona must have a convincing background, history and supporting details, and that up to 50 US-based controllers should be able to operate false identities from their workstations “without fear of being discovered by sophisticated adversaries”.
Centcom spokesman Commander Bill Speaks said: “The technology supports classified blogging activities on foreign-language websites to enable Centcom to counter violent extremist and enemy propaganda outside the US.”
He said none of the interventions would be in English, as it would be unlawful to “address US audiences” with such technology, and any English-language use of social media by Centcom was always clearly attributed. The languages in which the interventions are conducted include Arabic, Farsi, Urdu and Pashto.
Centcom said it was not targeting any US-based web sites, in English or any other language, and specifically said it was not targeting Facebook or Twitter.
Once developed, the software could allow US service personnel, working around the clock in one location, to respond to emerging online conversations with any number of co-ordinated messages, blogposts, chatroom posts and other interventions. Details of the contract suggest this location would be MacDill air force base near Tampa, Florida, home of US Special Operations Command.
Centcom’s contract requires for each controller the provision of one “virtual private server” located in the United States and others appearing to be outside the US to give the impression the fake personas are real people located in different parts of the world.
It also calls for “traffic mixing”, blending the persona controllers’ internet usage with the usage of people outside Centcom in a manner that must offer “excellent cover and powerful deniability”.
The multiple persona contract is thought to have been awarded as part of a programme called Operation Earnest Voice (OEV), which was first developed in Iraq as a psychological warfare weapon against the online presence of al-Qaida supporters and others ranged against coalition forces. Since then, OEV is reported to have expanded into a $200m programme and is thought to have been used against jihadists across Pakistan, Afghanistan and the Middle East.
OEV is seen by senior US commanders as a vital counter-terrorism and counter-radicalisation programme. In evidence to the US Senate’s armed services committee last year, General David Petraeus, then commander of Centcom, described the operation as an effort to “counter extremist ideology and propaganda and to ensure that credible voices in the region are heard”. He said the US military’s objective was to be “first with the truth”.
This month Petraeus’s successor, General James Mattis, told the same committee that OEV “supports all activities associated with degrading the enemy narrative, including web engagement and web-based product distribution capabilities”.
Centcom confirmed that the $2.76m contract was awarded to Ntrepid, a newly formed corporation registered in Los Angeles. It would not disclose whether the multiple persona project is already in operation or discuss any related contracts.
Nobody was available for comment at Ntrepid.
In his evidence to the Senate committee, Gen Mattis said: “OEV seeks to disrupt recruitment and training of suicide bombers; deny safe havens for our adversaries; and counter extremist ideology and propaganda.” He added that Centcom was working with “our coalition partners” to develop new techniques and tactics the US could use “to counter the adversary in the cyber domain”.
According to a report by the inspector general of the US defence department in Iraq, OEV was managed by the multinational forces rather than Centcom.
Asked whether any UK military personnel had been involved in OEV, Britain’s Ministry of Defence said it could find “no evidence”. The MoD refused to say whether it had been involved in the development of persona management programmes, saying: “We don’t comment on cyber capability.”
OEV was discussed last year at a gathering of electronic warfare specialists in Washington DC, where a senior Centcom officer told delegates that its purpose was to “communicate critical messages and to counter the propaganda of our adversaries”.
Persona management by the US military would face legal challenges if it were turned against citizens of the US, where a number of people engaged in sock puppetry have faced prosecution.
Last year a New York lawyer who impersonated a scholar was sentenced to jail after being convicted of “criminal impersonation” and identity theft.
It is unclear whether a persona management programme would contravene UK law. Legal experts say it could fall foul of the Forgery and Counterfeiting Act 1981, which states that “a person is guilty of forgery if he makes a false instrument, with the intention that he or another shall use it to induce somebody to accept it as genuine, and by reason of so accepting it to do or not to do some act to his own or any other person’s prejudice”. However, this would apply only if a website or social network could be shown to have suffered “prejudice” as a result.
• This article was amended on 18 March 2011 to remove references to Facebook and Twitter, introduced during the editing process, and to add a comment from Centcom, received after publication, that it is not targeting those sites.