Extremists “are using less and less phone lines and more and more internet connections,” said French Interior Minister Gerald Darmanin at a recent press conference. It makes a lot of sense: for one, everyone’s using less phone lines and more internet connections, and in addition, internet conversations can be harder to track down than phones.
After the terrorist attacks in 2015, France started a trial using algorithms to detect terrorist activities online. In 2017, the country passed a law allowing it to monitor instant messaging apps (though actually enforcing it is sometimes a challenge). But this was mostly experimental in nature until now.
The new bill would give France more power to detect potential terrorists, with Darmanin noting that intelligence services will, for instance, be able to spot someone who has accessed extremist websites several times.
This is, generally speaking, in line with global trends. Social networks like Facebook and Instagram are already using algorithms in the form of AI (artificial intelligence) to monitor hate speech and even drug dealers.
“Our technology is able to detect content that includes images of drugs and depicts the intent to sell with information such as price, phone numbers or usernames for other social media accounts,” explained Kevin Martin, head of US public policy at Facebook.
It only makes sense that counterterrorism follows suit. But the use of AI and big data, in general, have triggered intense debates between those who support the expansion of such technologies and those who see them as a threat to personal freedom.
“A new revolution has begun in counterterrorism—the Artificial Intelligence revolution,” one recent study noted. “The traditional delicate balance between effectiveness in the fight against terrorism and the liberal democratic values of society becomes even more crucial when counterterrorism engages in AI and big data technology.”
Concerns are further elevated by fears that algorithms just aren’t that good. Sure, the field of AI has been making strides in recent years, but there are still fears that for every detected terrorist, there could be 100,000 false positives.
Proponents of mass surveillance and this sort of detection algorithms invoke a distinction between collecting data and metadata with the idea that the latter is a lesser form of privacy violation. But algorithms aren’t that good at separating one from the other, and their results tend to be worse when they only work with metadata.
“Given the sparsity of datasets used for machine learning in counterterrorism and the privacy risks attendant with bulk data collection, policymakers and other relevant stakeholders should critically re-evaluate the likelihood of success of the algorithms and the collection of data on which they depend,” another recent study cautioned.
Still, even imperfect information can be useful, French authorities claim. The terrorist who killed a police employee in Rambouillet, south of Paris, a few days ago, had watched extremist videos just before carrying out his attack, a prosecutor said — and flagging this type of activity can save lives.
For France, the issue is somewhat political. Current president Emmanuel Macron is up for reelection soon, and his candidates are especially keen on attacking him on security. The far-right candidate Marine Le Pen who was Macron’s contender in last elections said French people were “encircled by delinquency and criminality”.
Since 2017, terrorist attacks in France have killed 25 people. Three quarters of these attacks were carried out by French nationals, prompting prime minister Jean Castex to emphasize that immigration policies and terrorism policies must be kept separate as they are two different things.
It’s hard to say just how good France’s algorithms are. If they’re anything like what research literature is discussing, then authorities will have to be careful to make sure they don’t end up causing more problems than they solve, but it’s not hard to understand the country’s desire to maximize security. Ultimately, France, like every other country, needs to decide how much of its freedom it wants to concede for the goal of security — a decision in which computers and algorithms will play an increasingly central role.