DARPA held a contest to identify evil propaganda robots on Facebook and Twitter
The US military has enlisted academics to fight a new enemy: Twitter bots.
The Defense Advanced Research Projects Agency (DARPA) held a special contest last year to identify so-called “influence bots” — “realistic, automated identities that illicitly shape discussion on sites like Twitter and Facebook.”
The fascinating 4-week competition, called the DARPA Twitter Bot Challenge, was detailed in a paper published this week.
The paper minces no words about how dangerous it is that human-like bots on social media can accelerate recruitment to organizations like ISIS, or grant governments the ability to spread misinformation to their people. Proven uses of influence bots in the wild are rare, the paper notes, but the threat is real.
And so, the surprisingly simple test. DARPA placed “39 pro-vaccination influence bots” onto a fake, Twitter-like social network. Importantly, competing teams didn’t know how many influence bots there were in total.
Teams from the University of Southern California, Indiana University, Georgia Tech, Sentimetrix, IBM, and Boston Fusion worked over the four weeks to find them all.
With 8.5% of all Twitter users being bots, per the company’s own metrics, it’s important to weed out those bots who go beyond just trying to sell you weight-loss plans and work-at-home methods, and cross the line into politics.
But actually making that distinction can be a challenge, as the paper notes.
Sentimetrix technically won the challenge, reporting 39 correct guesses and one false positive, a full six days before the end of the four-week contest period. But USC was the most accurate, going 39 for 39.
How to detect a robot
DARPA combined all the teams’ various approaches into a complicated 3-step process, all of which will need improved software support to get better and faster going forward:
- Initial bot detection — You can detect who’s a bot and who’s not by using language analysis to see who’s using statistically unnatural and bot-generated words and phrases. Using multiple hashtags in a post can also be a flag. Also, if you post to Twitter a lot, and consistently over the span of a 24-hour day, the chances you’re a bot go up.
- Clustering, outliers, and network analysis: That first step may only identify a few bots. But bots tend to follow bots, so you can use your initial findings to network out and get a good statistical sense of robot social circles.
- Classification/Outlier analysis: The more positives you find with the first two steps, the easier it is to extrapolate out and find the rest in a bunch.
A key finding from the DARPA paper, and very important to note, is that all of this required human interaction — computers just can’t tell a real human from an influence bot, at least not yet.
The good news, say the authors in their paper, is that these methods can also be used to find human-run propaganda and misinformation campaigns.
The bad news is that you can expect a lot more evil propaganda bots on Twitter in the years to come.
“Bot developers are becoming increasingly sophisticated. Over the next few years, we can expect a proliferation of social media influence bots as advertisers, criminals, politicians, nation states, terrorists, and others try to influence populations,” says the paper.
- This is the Marine Corps' first female infantry officer
- Russia just finished the Zapad military exercises that freaked out NATO — Here's what we know
- North Korea calls Trump's tweets a declaration of war, threatens to shoot down US bombers
- SOLDIERS SPEAK OUT ON KAEPERNICK: His protest 'makes him more American than anyone'
- A former Green Beret who comes from a military family lays out why he supports football players kneeling during the national anthem
- The Feds are looking into some suspicious Equifax trades (EFX)
Follow @BusinessInsider on Twitter .
Here's why the Russian navy just took a Philippine vacation
Russia has sent three antisubmarine ships to Manila in an effort to build stronger ties with and dissuade American ties from the Philippines.
This is why the Pentagon is investigating the ambush in Niger that killed 4 special operators
What exactly happened in Niger, and how did we lose four service members in a surprise ambush. There are questions, and McCain and Mattis want answers.
This South Korean howitzer can bring the thunder if Pyongyang attacks
The K9 can hit a target 25 miles away with three rounds in 15 seconds.
7 reasons why Obi-wan Kenobi was basically Ulysses S. Grant
Just replace Obi-wan's Spirit form at the end of Return of the Jedi with Grant's love of spirits and you could make a case for one inspiring the other.
This new technology can help tank crews 'see' through their armor
Being buttoned up in a tank used to mean being blind as a bat. With this new technology, that's no longer the case.
This is how John Kelly shut down speculation on President Trump's gold star family call
"If you're not in the family, if you've never worn the uniform, if you've never been in combat, you can't even imagine how to make that call," Kelly said.
Blumhouse and WATM team up to produce 'Searching for Bergdahl'
Blumhouse Television and WATM are teaming up to produce the documentary "Searching for Bergdahl," the untold story of the seven-year search for the missing soldier.
This is the real reason John McCain's Liberty Medal speech was so epic
When US media focused on a jab at President Trump, they missed the parting thoughts of a veteran and public servant of more than 60 years.
This little bot can take a lickin' and keep on tickin' for troops on assault
Weighing a little over five pounds, the FirstLook can handle being thrown into a hostile environment.