We must make it easier to prevent child abuse online

Meta's recent court defeat should spur governments around the world to step up their fight against online child abuse. A New Mexico court ruled that the company's platforms not only expose children to sexual predators and display sexually explicit material, but also mislead the public about these risks.

Despite this landmark decision, discussions about child safety online may continue to veer from one scandal to another without truly effective solutions being adopted. Public outcry usually peaks after major court cases or after shocking news stories, as was the case in January when it was discovered that the chatbot Grok was helping users flood social media with sexualized images of children. Politicians, of course, make noise, even promising strict regulation, but then public attention wanes and no action is taken. In the meantime, technology continues to evolve and create new problems.

It doesn’t have to be that way, because we already know how to prevent online violence. The problem is not a lack of knowledge. The problem is that our response is out of step with reality. A recent systematic review of the scientific literature and measures to combat violence in the real world, published by the International Panel of Experts on the Information Environment (IPIE), came to a devastating conclusion: while we already have the tools we need, the global response remains fragmented and incomplete.

Digital platforms and investigators, for example, rely heavily on systems that scan the web for known illegal images and detect “grooming” (lurking and gradually gaining the trust of the victim for abuse - ed.). Without these tools, many cases of online violence would go undetected. Artificial intelligence is increasingly being used to improve these tools and analyze vast amounts of data from online platforms, allowing investigators to find previously undetected material with elements of violence.

However, the effectiveness of AI depends on the data used to train the models, and experts often do not have reliable access to the data sets they need. The authors of the IPIE analysis call on governments to create a safe, controlled environment in which authorized experts can review sensitive material, while ensuring the protection of victims. Without such cooperation, progress will remain uneven and unstable.

Moreover, all of these methods are aimed at combating violence after it has occurred. While it is important to improve methods for detecting perpetrators, the goal should be to prevent harm before it occurs. Here, the IPIE team uncovered important “blind spots.” To organize and profit from sexual violence, perpetrators can exploit digital payment systems, advertising tools, and other widely used features of online platforms. However, the measures taken rarely target these mechanisms. If the financial infrastructure that supports sexual exploitation remains intact, the fight against violence will always lag behind.

In addition, the public must be informed. Educational campaigns that warn minors about the risks of grooming and explain the harmfulness of illegal content show encouraging results (behavior becomes safer, the number of reported violations increases), but this work is still insufficiently active. In any case, education cannot replace responsible operation of platforms or effective law enforcement. Children should not be asked to use systems that allow them to be exploited. Responsibility must lie with the companies that create and operate such platforms.

Although big tech companies operate globally, law enforcement remains largely national. Without a strengthened legal framework that applies internationally, abusers will exploit regulatory gaps. In some countries, authorities already have the power to enforce laws beyond their national borders, requiring companies to report abuse, preserve evidence, and cooperate with investigators. The European Union, one of the largest markets for all major digital platforms, is, for example, particularly well-positioned to hold both the largest platforms and criminals accountable.

However, as IPIE’s analysis shows, much more can be done to improve international coordination to ensure that enforcement of laws against companies and criminals is clearer and more consistent. The Grok scandal is a clear example of the difficulties regulators face in keeping up with rapidly evolving technologies. While regulators reacted quickly when it became clear that a chatbot was creating child sexual abuse material, it took a serious public outcry to draw attention to the problem. In other cases, lobbying, economic pressure, and other forms of industry influence have prevented authorities from taking serious regulatory action.

Taken together, the IPIE findings reflect the state of global measures to combat online child abuse: they are absolutely necessary, but they are insufficient. Detection tools regularly identify cases of abuse that they have been trained to detect; law enforcement agencies prosecute perpetrators where possible; educational programs try to draw attention to the ongoing risks. But the entire system still struggles to keep pace with the ever-changing digital landscape. The financial systems that support sexual exploitation remain poorly regulated; investigators lack access to the data needed to improve the detection process; and legal frameworks fail to take into account the global nature of the problem.

For years, governments have argued that more data is needed for more decisive action. This justification is no longer convincing. As the IPIE team has found, the data is there. We already know which measures are helping to reduce harm, and we know where the key gaps are. The only unknown that remains is whether political leaders will take the necessary steps to ensure children’s safety.

The author is Professor of Digital Rights at Bournemouth University and Visiting Professor at the University of Suffolk.

Copyright: Project Syndicate, 2026. (translation: NR)