Twitter, Inc. v. Taamneh
The Supreme Court unanimously ruled that Twitter, Facebook, and Google cannot be held liable for a deadly 2017 ISIS attack in Istanbul, because simply allowing terrorists to use publicly available platforms — without any targeted, intentional assistance directed at the attack — does not qualify as aiding and abetting under federal law.
The decision draws an important line for social media companies: knowing that bad actors use your platform, and failing to remove them quickly enough, is not the same as consciously helping them carry out violence.
How it got here: A federal trial court dismissed the families' complaint; the Ninth Circuit reversed and ruled the families had stated a valid claim; Twitter petitioned the Supreme Court, which agreed to hear the case.
The Case in Depth
What happened
On New Year's Eve 2017, an ISIS operative killed 39 people and wounded dozens more at the Reina nightclub in Istanbul, Turkey. The family of one victim sued Twitter, Facebook, and Google, arguing those companies had knowingly allowed ISIS to use their platforms for years to recruit members, raise money, and spread propaganda — and that their recommendation algorithms actively directed ISIS content to receptive users — making the companies responsible for the attack under a federal law allowing civil suits against those who aid and abet international terrorism.
The question before the Court
Can victims of a terrorist attack sue Twitter, Facebook, and Google for civil damages, arguing the companies aided and abetted the attack by knowingly allowing ISIS to use their platforms for years?
The Court's answer
No — the families' allegations were not enough to state a claim that the social media companies aided and abetted ISIS in carrying out the Istanbul nightclub attack. The federal law at issue (JASTA) requires "knowing and substantial assistance" to the actual terrorist act, grounded in the common-law idea that aiding-and-abetting liability falls only on those who consciously and culpably participated in the wrongdoing to help it succeed. The companies gave ISIS the same passive, algorithm-driven access they gave billions of other users; they never singled out ISIS for special help, and no allegations connected their platforms to the planning or execution of the specific attack.
Because the companies' alleged failing was mostly passive — not removing ISIS content promptly rather than actively helping ISIS — the families needed to show a strong, direct link between the companies' services and the attack. They could not do so. Accepting their theory would expose any communications provider to liability for every crime committed by anyone using its services, which would stretch aiding-and-abetting law far beyond its traditional limits.
Curious how the Court got there? See the step-by-step legal reasoning →
Why it matters
Social media companies and other internet platforms now have clearer protection against federal terrorism lawsuits that rest solely on the theory that terrorists used their services. To face liability, a company would need to have done far more — actively and intentionally helping plan or carry out a specific attack, not merely offering the same general platform available to billions of users worldwide.
What changes now
The Ninth Circuit's ruling for the families is reversed and the complaint is dismissed. The families cannot recover from the social media companies on these allegations. The ruling applies alongside the companion case, Gonzalez v. Google, involving nearly identical claims. Future lawsuits with different facts — for example, a platform that actively and selectively promoted specific terrorist content, or provided unusual targeted services to a known terrorist group — could potentially fare differently under the framework the Court laid out, but the Court left those questions open.
What this does not decide
The Court did not decide whether a social media platform that consciously and selectively promotes terrorist content, or that provides unusual or targeted services to a terrorist group beyond general availability, could face liability. The ruling is limited to these specific allegations, which showed no meaningful distinction between ISIS's use of the platforms and any ordinary user's.
Concurrences and dissents
Concurrence — Justice Jackson
Justice Jackson joined the majority opinion but wrote separately to stress that the ruling is narrow in important ways. She emphasized that the Court's characterization of the social media platforms and algorithms rests entirely on the specific allegations in these complaints, filed at the motion-to-dismiss stage with no factual record, and that cases with different allegations and a developed record might come out differently. She also cautioned that the common-law principles the Court used to interpret JASTA do not automatically apply in other legal contexts.
How the Court got there
The legal reasoning, step by step
- JASTA imposes civil liability on anyone who 'aids and abets, by knowingly providing substantial assistance' an act of international terrorism, but defines none of those terms. Because 'aids and abets' is a familiar common-law phrase, the Court applied the principle that such terms 'bring the old soil' with them — meaning the Court looks to centuries of common-law meaning. Congress also pointed directly to a 1983 D.C. Circuit case, Halberstam v. Welch, as the proper legal framework for civil aiding-and-abetting liability.
- Halberstam requires three things: (1) the primary wrongdoer committed an act causing injury; (2) the defendant was generally aware it was playing some role in an overall illegal scheme; and (3) the defendant knowingly and substantially assisted the primary wrongdoer. Six factors help assess whether assistance was 'substantial,' including its nature, amount, duration, the relationship between the parties, and the defendant's state of mind.
- The core principle animating aiding-and-abetting liability is culpability: a defendant must have consciously and voluntarily participated in the wrongdoing in a way designed to help it succeed. The 'knowing' and 'substantial' parts of the test work in tandem — less direct assistance demands a stronger showing of intent, and vice versa. Passive bystanders or those providing widely available services without intending to assist a specific wrong do not qualify.
- The families cleared the first two Halberstam elements — ISIS committed the attack and the companies knew ISIS was using their platforms. The key dispute was the third element: whether the companies gave knowing, substantial assistance specifically to the Reina nightclub attack. The companies' recommendation algorithms, though active in sorting content, were agnostic to content type and treated ISIS the same as all other users, with no special encouragement or targeted help directed at ISIS.
- Because the alleged failing was largely passive nonfeasance — not removing ISIS rather than actively helping it — the families needed a strong showing of intentional, substantial assistance. That was absent: no allegations tied the platforms to the specific planning of the attack, no special treatment was given to ISIS, and the companies' relationship with ISIS was arm's-length and indifferent, just like their relationship with billions of other users.
- The Court also rejected any theory that hosting ISIS content generally could make the companies liable for every ISIS attack worldwide. Such sweeping liability would require a showing of truly pervasive, systemic, and intentional assistance to ISIS as an enterprise — far beyond passively hosting content — and would otherwise reduce any communications provider to an insurer against all crimes its users commit.
Doctrinal impact
Cases affected by this decision
Reaffirms Halberstam v. Welch (705 F. 2d 472)
Confirmed as the governing legal framework for civil aiding-and-abetting liability under JASTA, though adapted to new facts.