Billions of everyday citizens are using the social media platforms. So are the terrorists. YouTube is used to distribute videos of radical content. Facebook is used for operational and tactical information, such as bomb recipes, weapon maintenance, tactical shootings, links to extremist sites and Facebook groups according to the US Department of Homeland security. Twitter has been used for recruitment and radicalisation by the Islamic State.

Tech companies have been tackling the issue of fighting online terrorist content proactively using artificial intelligence for flagging terrorist content. In September 2017, Twitter announced in its transparency report the taking down of 300,000 terrorist accounts in that year only – three quarters of them even before the tweet was sent. As well as YouTube previously launching an experiment to fight the propaganda-related content against ISIS recruits and Facebook has taken responsibility in eliminating online terrorist content from its platform.

As terrorist attacks have been proliferating, US and European governments are pushing the responsibility on the tech companies and urging them to take countermeasures against terrorist content spreading on social media platforms. In 2016 internet companies, such as Google, Youtube, Facebook and Twitter, aligned with the European Commission by applying technology to automatically detect terrorist material.

However the government wants more. In 2017, Theresa May, the Prime Minister of the United Kingdom urged Google, Facebook and Twitter to take down terrorist content within two hours or face heavy fines. She made this announcement after meeting French President Emmanuel Macron, and Paolo Gentiloni, the Prime Minister of Italy at the general assembly of the United Nations. May added that “the industry needs to go further and faster in automating the detection and removal of terrorist content online, and develop technological solutions which prevent it being uploaded in the first place.” Technologies such as the INVISO Intelligence Platform which helps law enforcement agencies detect radical content online through the use of data mining and natural language processing.

Such collateral damage are the false positives, when artificial intelligence-based algorithms of the big internet companies detect and take down content which were not meant to be terrorism-related and gets removed accidentally. Such automatised taking down of content does harm freedom of speech.

However, terrorists are not only using the big internet companies’ platforms, but also more and more the smaller ones using end-to-end encryption, such as Telegram, WhatsApp or Snapchat. Telegram is used by ISIS to share and disseminate propaganda, but the platform also fights. The attacker at London bridge in 2017 sent his last message using Whatsapp, and police had a hard time in getting the content of the message. Members of DAESH behind the 2015 Paris attacks were also using Telegram to spread propaganda, and also used it to recruit people for the Christmas market attack in Berlin in 2016. Snapchat is popular among jihadist extremists too.

Even if big internet companies manage to shut down online terrorist content from their platforms using sophisticated algorithms, the problem itself will not cease to exist. Terrorists will find their ways to communicate via the internet, either by the enlisted end-to-end encrypted social media platforms, or others unknown as of now. Governments and companies need to adapt to the ever-changing situation using an evolving technology – but also keeping the privacy of the citizens in the forefront of the fight against terrorism.