At the end of November, the member states of the Council of the European Union (EU Council) approved a compromise proposal for Chat Control regulation. Slovakia voted against this proposal in the EU Council on the grounds of privacy protection and the risk of setting a precedent for mass monitoring of communications.
The approved proposal on protecting children from abuse by sexual predators on the internet refers to voluntary scanning of messages, not mandatory scanning of all users. However, the proposal still has to be reviewed by the European Parliament.
Artificial intelligence (AI) is increasingly interfering with our lives. Young people are beginning to take it for granted as a natural part of their lives. What do you know about minors in relation to artificial intelligence?
Children today are exposed to risks in the online space that did not exist a few years ago. Social networks have a very negative impact on children's mental health, but psychologists are finding, as confirmed by experts from IPčko, for example, that children refer to AI as their "best friend."
Anyone who has ever used artificial intelligence such as ChatGPT, Grok, and the like has noticed that the robot agrees with them on everything, confirms their answers, and never questions anything, which children like.
But the real world is also about disagreement, about coping with crisis and conflict situations, about encountering other opinions. The fact that minors are forming relationships with computers is truly frightening.
There is recent talk of banning and regulating social media for children and adolescents. Are we really in such a bad state?
Yes, we are. Children are addicted to social media. Social media has addictive content [causing addiction, ed.] and, of course, content that presents them with a perfect life, literally unrealistic, deceptive, and what appears to be organic content is just a new form of advertising.
All of this is heavily promoted by influencers who are paid for advertising. Let's also remember the various "cool challenges" that minors try to overcome, and we have a complete online disaster. Eating disorders, health damage, injuries, but unfortunately also cyberbullying, suicide attempts, self-harm, non-substance addictions...
Do you have any specific examples from your research?
When we visited schools last year for various meetings and lectures, often as part of the Anti-Bullying Academy, where we talked directly with children about the phenomenon of cyberbullying, we asked the children to raise their hands if they had ever been approached by a stranger on any platform on the internet.
There are always about 80 raised hands, which is a frightening number. I often see cases of cyberbullying, blackmail, the dissemination of photos of children, or grooming [attempts by adults to establish contact with minors for the purpose of sexual abuse, ed.].
You also warn minors about addictive substances.
There is also financial and property fraud, not to mention that we recently discovered in our own investigation how easily young people can access nicotine products through social networks, for example through Facebook Marketplace or the largest Slovak online bazaar. Private individuals sell vape or nicotine pouches there completely unregulated.
We also have information from children about being directly approached via Instagram messages with offers of nicotine products.
Protecting children in the online space is therefore appropriate.
Protecting children from sexual predators and any kind of abuse, including economic abuse, is an absolute priority. Children are not yet able to assess the risks of the online world and cannot think through the consequences. This is the task of us adults, and above all, it is a parental responsibility.
But let's be honest, how often are it the parents who give their child their first smartphone and allow them to register on social networks before they reach the age of 13, with the excuse that "all their friends are there anyway"?
The European Union has come up with the Chat Control proposal, whereby artificial intelligence would monitor private conversations in order to protect minors from sexual abuse. What do you think about this?
Every good idea of this type has its risks, which need to be carefully considered and discussed with experts. There are several types of artificial intelligence on the market, some more popular than others, and logically, each one belongs to someone, each one was created by some IT giant.
The question is: is it good for a private company to have access to complete conversations? After the EU implemented the GDPR, will we allow someone to read all conversations in their entirety? Can anyone guarantee that there will be no information leaks these days?
That is why it is necessary to look at this seemingly ingenious solution from a multi-faceted perspective.
Do you have any examples of minors who have encountered a certain type of sexual abuse? I am referring to sexual predators, pedophiles.
As soon as we receive such a report, which constitutes a criminal offense, we notify the law enforcement authorities.
I have several expert advisors and colleagues who encounter all kinds of things, but I must emphasize that this is in the offline world. For example, at a recent press conference in Prešov as part of our field monitoring of begging children, we presented the findings of colleagues who uncovered not only organized begging, but also sexual abuse and prostitution of minors.
And unfortunately, as the children themselves told us, the customers are from the white majority and are wealthy people. We have handed this over to the police.
The Chat Control proposal is welcomed mainly by children's organizations that are in favor of greater protection for children on the internet. It would also help the police, for example, to track down videos and photos related to cyberbullying that are shared among children's peers. How does your office view this?
First of all, it must be said that protecting children on the internet starts with parental responsibility. If that worked, we wouldn't have children under the age of 13 on social networks today.
Registration can be easily circumvented.
The problem is that although social networks are now set up so that only people over the age of 13 can register, in reality this is only a unilateral statement that can be bypassed, and there are significantly younger children on social networks.
They are thus exposed to content that is not suitable for them, or even harmful to their mental health.
What do you know about minors on social networks?
The reality? When we surveyed 13-year-olds, 72 percent of them considered it important to have access to the internet, and as many as 40 percent expressed great dissatisfaction with the number of followers they had. I am primarily in favor of restricting children's access to social networks.
The goal of monitoring private communications is for artificial intelligence to search for material depicting child sexual abuse. However, there is talk that algorithms may incorrectly flag innocent content. Artificial intelligence may therefore also flag photos sent between adult couples with a slightly "sexual charge" as child sexual abuse. Do children also share such "innocent content"?
From conversations with experts, teachers, and children, we know that they do send such content to each other. And that parents would be very surprised.
As for AI's ability to detect harmful content, I ask again whether we can guarantee its accuracy, given that we already know what fatal mistakes it makes even with common factual questions.
Another question is why we want to expand the capabilities of AI, rather than limiting its use and capabilities. What about deepfake videos, for example, which can create fake compromising material about you? Someone could blackmail you with something created by artificial intelligence.
There are also critical voices saying that this is a violation of privacy in a way, as it is a form of surveillance. What is your position on this? Isn't it unconstitutional in a sense?
The competent authorities must comment on the unconstitutionality. A broader professional discussion on this issue is needed.
In your opinion, is there any other way to prevent child abuse in the online space without having to monitor private communications?
Restricting children's access to social networks, limiting the content that is displayed to children, and strengthening the relationship between parents and children in terms of building greater trust, talking about risks, and prevention. This starts at home with a strong parent-child relationship.
