A child sitting alone in a bedroom can become the starting point for an international criminal investigation. That is the lesson from the arrest this week of a 16-year-old in Malmesbury, in the Western Cape, in connection with a loose collection of online networks known as 764.
The networks target vulnerable children through social media, gaming platforms and messaging services. According to the FBI, perpetrators can begin by posing as friends before using manipulation, threats and blackmail to coerce victims into producing sexual material or engaging in self-harm, violence or animal cruelty.
South African police arrested the teenager on August 17 after intelligence from the FBI, the US Embassy and the National Center for Missing & Exploited Children pointed investigators towards a person suspected of links to 764. The allegations against the teenager have not been tested in court.
The striking feature of the case is not simply the alleged conduct, but how many borders the investigation has already crossed.
The suspect is in South Africa. The intelligence came partly from America. The alleged activity took place online. The platforms involved operate internationally. And evidence in such investigations can be scattered across devices, companies and jurisdictions.
The result is a policing problem in which the victim, suspect, platform and evidence may all sit in different jurisdictions..
A network without a headquarters
The FBI does not describe 764 as a conventional organisation with a fixed membership and hierarchy. It uses the term as a catch-all for diffuse networks and offshoots whose names can change while their methods remain similar.
Their targets are often minors or other vulnerable people.
The FBI says perpetrators may find potential victims on gaming platforms, social media and messaging applications. They can spend time building trust before turning that relationship into leverage.
Once a victim has shared something compromising, the dynamic can change. Threats to expose material, harm the victim or target relatives can be used to obtain further material or force the victim to carry out increasingly extreme acts.
That creates an unusual criminal economy: the victim can be made to produce the evidence of their own exploitation.
The FBI warned in March 2025 of a sharp increase in activity by 764 and related networks operating in America and around the world.
By February this year, the FBI's Boston division said it was investigating more than 350 subjects nationwide tied to networks commonly referred to as 764. It warned that the label covers numerous offshoots.
Traditional policing is built around organisations, places and jurisdictions, while online networks are not.
America's response is getting tougher
The United States has moved beyond simply warning parents. Federal prosecutors have charged and convicted alleged members of 764 and related networks in cases involving child exploitation, cyberstalking and violence.
In March, Erik Lee Madison pleaded guilty after US authorities said he had sexually exploited at least 10 minor girls and used threats and blackmail to compel victims into self-harm and other acts.
In June, Shawn Krishendat Premsook pleaded guilty to child-exploitation offences. Prosecutors said he had participated in extorting minors into self-harm and described 764 as a nihilistic violent-extremist network operating in America and abroad.
In July, another alleged member of a related network was sentenced to 40 years in prison. The Justice Department said these networks can operate internationally and target vulnerable people through social media.
US authorities are increasingly seeing an overlap between online exploitation, cybercrime and violent extremism.
South Africa has now encountered that problem directly.
The jurisdiction problem
The internet has made it possible for a crime to be simultaneously local and international.
A child can be sitting in Cape Town or Malmesbury. The person communicating with them might be somewhere else. The platform might be headquartered in another country. The relevant data might be stored elsewhere again.
For investigators, each step can create a different legal and technical problem.
Who has the evidence?
Which country's laws apply?
How quickly can digital evidence be obtained before it disappears?
How easily can investigators identify the person behind an account?
And how much cooperation can foreign agencies provide to a South African investigation?
The Malmesbury case offers a partial answer to the last question: international intelligence-sharing is already part of the response.
But intelligence-sharing is not the same as having jurisdiction. That distinction will become increasingly important as criminal networks use global platforms to operate across borders.
The challenge for South African law enforcement is not simply finding offenders. It is building the technical expertise, international relationships and investigative capacity required to follow a crime that does not stay inside the country where its victim lives.
That may be the more important lesson from the Malmesbury investigation. The ability to exchange intelligence is one thing; the ability to investigate, obtain evidence and prosecute across jurisdictions is another.
Parents are part of the security system
There is another uncomfortable implication. Police may ultimately be able to identify an offender. But they cannot monitor every private conversation between a child and a stranger online.
That makes parents an important early-warning system.
The FBI has issued unusually direct guidance to parents and caregivers. It says children may be approached through gaming, messaging and social-media platforms and that perpetrators may deliberately cultivate friendships or romantic relationships before beginning coercion.
The warning signs may therefore be behavioural rather than technological.
A child may suddenly become secretive about their phone. Their sleep or eating habits may change. They may become withdrawn or unusually irritable. They may appear frightened about something happening online or become desperate to conceal a relationship.
None of these signs proves that a child is being targeted, but they can provide a reason to ask questions.
And how parents ask those questions matters. A child who believes that admitting they have sent an image, spoken to a stranger or been threatened will result in their phone being confiscated may decide that silence is safer.
The FBI's advice emphasises communication rather than simply restricting access.
The bigger lesson
The South African case demonstrates what happens when that phenomenon encounters a national police force: the investigation immediately becomes international.
The internet has made geography less of a barrier to criminal networks. A person looking for a vulnerable child no longer needs to enter the child's neighbourhood, school or city.
For police, catching such offenders increasingly depends on cooperation between countries and companies. Meanwhile for parents, the response is more immediate: knowing enough about their children's online lives to recognise when something is wrong, and making sure a frightened child knows whom to tell.
The response is also shifting beyond the family. OpenAI, for example, has introduced stronger safeguards for teen accounts, parental controls, quiet hours and limited safety notifications. Parents who link accounts can manage selected settings, although they cannot read their teenagers' conversations.
That points to a broader change in online safety: parents remain an important line of defence, but they are increasingly being joined by the platforms themselves, which are being asked to build more protection into the technology children use.
What parents can actually do
The practical response is less about trying to make the internet disappear and more about making children harder to isolate and manipulate.
- Know the platforms your child uses. Gaming services can be as important as social-media applications.
- Ask about online friendships. Not as an interrogation, but as part of ordinary family conversation.
- Explain that identities can be false. A person who appears to be another teenager may not be one.
- Talk about intimate images. Children should understand that once an image is sent, they may lose control over it.
- Take secrecy seriously. Particularly when someone online insists that a relationship or conversation must remain hidden.
- Watch for changes in behaviour. Withdrawal, unusual secrecy, disrupted sleep or sudden distress can warrant a conversation.
- Take threats seriously. Requests involving self-harm, violence, sexual material or animal cruelty should never be dismissed as a joke.
- Don't make the child feel like the offender. Manipulation works partly because victims are frightened of the consequences of disclosure.
- Preserve evidence. Keep relevant usernames, messages and other digital information if an incident is reported.
- Report suspected exploitation. Parents should not attempt to confront suspected offenders themselves.