In late 2015, candidate Donald Trump announced he would thwart the Islamic State by cutting its access to the internet.

"We're losing a lot of people because of the internet and we have to do something," he said during a rally in South Carolina. "We have to go see Bill Gates and a lot of different people that really understand what's happening. We have to talk to them, maybe in certain areas closing that internet up in some way."

Trump's promise came amid a rash of attacks around the world; a month before his rally, ISIS carried out a series of coordinated attacks in Paris, killing 130 people.

Two years later, as president, Trump doubled down on his goal, saying the internet was the terrorist organization's "main recruitment tool which we must cut off & use better!"

Today, ISIS has lost its territorial strongholds and is greatly diminished — but not because Trump stopped it from using the internet.

Unlike other countries, the United States does not require internet service providers or social media platforms to remove terrorist content. Federal law generally says internet service providers aren't legally responsible for what users post on their platforms. The U.S. also hasn't signed onto international, non-binding pledges to remove terrorist content online.

"The United States believes that protecting human rights and fundamental freedoms, including freedom of expression, is an important part of our counterterrorism strategy because increased censorship and other restrictions on human rights can lead to greater instances of terrorist radicalization," the State Department wrote in its 2019 terrorism report.

Much of the progress that's been made to reduce the influence of ISIS online comes from tech companies like Twitter, Facebook and Google, which automatically remove what they consider terrorist content from their platforms. Still, Trump tried to open internet platforms up to more liability.

In May 2020, Trump signed an executive order calling for limitations on Section 230(c) of the Communications Decency Act, the provision that grants tech platforms their legal immunity. In December, he vetoed a military spending bill in part because lawmakers did not include a repeal of Section 230. (Congress voted to override his veto, a first in the Trump years, and the measure became law.)

Trump's issue with Section 230 wasn't that it gives cover for ISIS to continue recruiting on the internet — rather, he said he was concerned that it let tech platforms stifle conservative opinions. 

Congress hasn't repealed Section 230, nor has it passed legislation to close parts of the internet to terrorist organizations. Experts previously told PolitiFact that wouldn't be feasible, given how the internet works and how the First Amendment protects speech.

So, where does that leave ISIS?

The United Nations says more than 10,000 fighters are active in Iraq and Syria, where attacks are on the rise. The internet "remains a serious tool for terrorism and extremism that goes far beyond national boundaries," according to the Center for Strategic and International Studies. Intelligence groups have said the coronavirus pandemic opened up more avenues for terrorist groups to recruit online.

We rate this Promise Broken.

Donald Trump's campaign promise to close parts of the internet to the Islamic State has not materialized. However, the depletion of the terrorist organization's territory, as well as moderation on social media platforms and international law enforcement action, has resulted in a decline in its online activity during his presidency.

In December 2015, Trump called for "closing that internet up in some way" to impede the recruitment of ISIS fighters. He floated the idea of having Bill Gates, the co-founder of Microsoft, look into the possibility.

Nearly five years later, there have been several successes in the effort to defeat ISIS. By January 2018, the terrorist organization had lost 93% of its territory in the Middle East. In October 2019, Abu Bakr al-Baghdadi, the leader of ISIS, died in a raid by U.S. forces.

Still, ISIS has thousands of fighters and continues to recruit and publish propaganda online. Unlike other countries, Congress has not yet passed legislation that would require internet service providers or social media platforms to remove terrorist propaganda (although some legislators have introduced bills that would address it).

But many of them have done so, anyway.

Since Trump's election, artificial intelligence has helped social media platforms remove content from terrorist organizations. Facebook has reported that it automatically removes more than 99% of ISIS and al-Qaida content published on its platform. YouTube reported that, between October and December 2019, about 90% of videos that violated the company's policy against violent extremism were removed before they had 10 views.

In May 2019, prompted by the terrorist attack at mosques in Christchurch, New Zealand — which was live-streamed on Facebook — several governments and online service providers adopted a non-binding pledge called the "Christchurch Call to Action to Eliminate Terrorist and Violent Extremist Content Online." Tech companies like Twitter, Facebook and Google pledged to take additional measures to address terrorist content on their platforms, such as updating terms of service and improving technology that detects terrorist content.

Although it has endorsed similar intergovernmental initiatives, the U.S. did not sign the pledge "due to policy and legal concerns," according to a 2019 State Department terrorism report. 

"It is true that the big tech platforms have somewhat cleaned up their acts in this regard, most especially Twitter," said Cori Dauber, a communication professor at the University of North Carolina at Chapel Hill, in an email. "But none of that, as far as I know, is a function of federal action — the platforms were responding to pressure, sure, but they've been under pressure on these issues for years."

Even if the U.S. did pass legislation that forced internet platforms to ban or block terrorist propaganda, the First Amendment could stand in the way. Federal law also generally protects internet service providers from legal responsibility for what users do or say on their platforms, which is different from European countries like France, which has passed legislation forcing tech platforms to remove terrorist content.

Trump has tried to challenge those protections — not to counteract terrorists but as a threat to ensure that conservative voices aren't silenced by content moderators. In a May 2020 executive order, the president called for limitations on Section 230(c) of the Communications Decency Act, the law that grants tech platforms their legal immunity. The order argues that, if platforms restrict certain voices, their protections should be revoked. Law experts have said that, absent legislation from Congress, the order would not be legally binding.

Finally, even if Trump did manage to ban ISIS from certain parts of the internet, it's unlikely that would stop the organization's recruitment.

In April 2018, authorities in eight countries took part in an operation that seized ISIS servers that hosted some of its propaganda apparatus. A year and a half later, Interpol, an international police organization in Europe, announced that it had stripped ISIS propaganda from platforms like Twitter and Google. Telegram, an encrypted messaging service, contained the most offending material.

However, shortly following the takedown, the BBC reported that ISIS supporters were talking about moving to other, more underground platforms to evade authorities and promote propaganda. And ISIS is known to use the darknet, parts of the internet that are anonymized and not searchable through traditional search engines like Google.

"Going after terrorist communications networks is kind of like treating a gunshot wound with a Band-Aid," said Patrick Eddington, a research fellow at the Cato Institute. "The real key to defeating them is defeating the ideology."

Despite American tech companies' in-house efforts to moderate and remove terrorist content, as well as international efforts to take down propaganda, it's clear that ISIS is still using parts of the internet to publish propaganda and recruit new members. And the U.S. government has not passed legislation that compels internet service providers to ban content from terrorist organizations.

"You can't really 'shut down' or 'close down' the internet," said Chelsea Daymon, a terrorism and political violence researcher at American University. "While you can remove content and shut down accounts that are pro-ISIS (which will slow the spread of content reach), the concept of shutting down parts of the internet shows a misunderstanding about what the internet is and how it works."

We rate this promise Stalled.

Shortly after news broke of the Sept. 15 Parsons Green bombing in London, President Donald Trump reiterated his campaign promise to close down portions of the internet to curb the spread of terrorism.

"Loser terrorists must be dealt with in a much tougher manner," Trump said in a tweet. "The internet is their main recruitment tool which we must cut off & use better!"

Prior to the election, he suggested U.S. antiterrorism officials would work with leaders in the technology industry to shut down parts of the internet that terrorist groups such as ISIS, or ISIL, use to recruit and share information.

U.K. Prime Minister Theresa May shared similar views in June after a different terrorist attack occurred on the London Bridge. Google, Twitter, Facebook and Microsoft responded to criticisms by vowing to improve detection of extremist content.

But cracking down on unwelcome or violent speech has proven to be difficult for those companies. A ProPublica investigation released in June found that Facebook's algorithm for identifying hate speech wound up removing some posts for nonsensical reasons and allowing other posts that should have been removed. Likewise, Twitter says it has shut down nearly 1 million suspected terrorist accounts since August 2015, but the company has struggled to find the "magic algorithm" to identifying terrorist content.

However, crackdowns on terrorist-related content appear to merely shift the problem elsewhere. Recent research shows that suspected terrorists have moved their online presence from Twitter to Telegram, a messaging app that was created in 2013.

The extent of the U.S. government's power to shut down online terrorist propaganda is unclear. When former Federal Communications Commission Chairman Tom Wheeler was questioned on the topic in November 2015, days after a mass shooting took place at a concert in Paris, Wheeler responded that the commission doesn't have the legal authority to shut down individual websites.

Even if legislation were passed to ban or block terrorist propaganda online, First Amendment rights could stand in the way. In addition, United States law generally protects internet service providers from legal responsibility for what the companies' users say or do on the internet, a marked difference from European countries such as France, which holds internet service providers responsible for removing terrorist propaganda.

Even though internet providers and websites have certain rights, there are still legal methods of taking down online content, according to Daphne Keller, director of intermediary liability at the Stanford Center for Internet and Society. U.S. Immigrations and Customs Enforcement has taken some of those legal avenues in the past to remove sites with child sex abuse material or with large-scale piracy.

"These kinds of blocks have been condemned by human rights bodies because they are poorly targeted and often take down legal speech with the illegal," she said. "I have not heard of this being done for terrorism, though it might be."

There might also be legal ground to compel third-party platforms, such as Twitter and YouTube, to remove content uploaded by known terrorists and terrorist organizations, but Keller said this has not yet been court-tested.

Meanwhile, Secretary of State Rex Tillerson recently accepted $60 million from Congress to continue the United States' counter-Russian and terrorist propaganda efforts (after sitting on the funding as lawmakers pressured him to take it).

For Trump's part, the president signed a joint statement among G7 leaders in May to put pressure on technology companies to eliminate online extremism.

Trump has reiterated this promise as president with subsequent terror attacks, but the chances of shutting down the internet don't look great in the United States based on stumbling efforts so far by tech companies and the First Amendment's speech protections. We rate this promise Stalled.