- Grok AI generated 6,700 sexually suggestive images per hour on X in a January 2026 24-hour monitoring test – 85 times more than the next five platforms combined.
- The European Commission has opened a formal investigation into X's Grok AI tool following the monitoring.
- Three months after Australia's under-16 media ban, 70% of affected users still had access to their Instagram, Snapchat, and TikTok accounts.
- European teens average 4.5 hours online on school days and over 6 on weekends, according to Eurobarometer.
- EIGE recommends mandatory, gender-responsive digital literacy across primary and secondary education, co-designed with young people.
A new chapter in protecting children online
Europe's debate on children's safety on social media has intensified, with the central question being whether the answer lies in digital literacy or outright media bans. EIGE's (European Institute for Gender Equality) latest research across ten Member States indicates that the prevailing approach – talking to students about online safety only after an incident – is no longer sufficient in a time when artificial intelligence has fundamentally changed the nature of cyber abuse.
AI has changed the threat
In a 24-hour monitoring window in early January 2026, researchers recorded that X's Grok AI tool generated approximately 6,700 sexually suggestive or nudifying images per hour. The next five largest producers of similar content generated just 79 images per hour combined – meaning Grok's output was around 85 times higher.
"We cannot wait until something goes wrong to start talking to children about life online. Digital literacy must be part of their education from the start, giving them the tools to navigate the online world safely and confidently," says EIGE Director Carlien Scheele.
The European Commission has since opened a formal investigation into the tool. What distinguishes AI-generated abuse from earlier forms of cyber crime is that no real photograph is needed. A school portrait or a family snap lifted from a parent's social media profile is sufficient to create an unwanted image. The violation is no longer about consent to share an image, but about consent to exist as one at all.
Teens spend half their waking hours online
Adolescents in Europe now spend an average of four and a half hours a day online during the school week and over six hours at weekends, according to Eurobarometer. Most of that time is invisible to the adults around them.
EIGE researchers conducted focus groups with girls aged 13 to 18 and boys aged 15 to 18 across ten Member States as part of their work on gender-based cyber violence. The conclusion: prevention and education systems across Europe do not reflect the digital realities young people are living in. Schools tend to address online safety after an incident, and when young people do report what is happening to them, adults often minimise it, dismiss it, or break their confidence.
Evidence against blanket media bans
Much of the current political energy in Europe is going into age verification, minimum ages, and national bans modelled on Australia's. Two thirds of Europeans now want EU rules limiting children's access by age.
The Lisbon Council's Evidence Hub compiles figures that nuance the debate:
| Indicator | Result |
|---|---|
| 15-year-olds who can manage digital privacy settings | 51% |
| Platforms informing children that they will investigate a report | Increased from 13% to 27% |
| Actual removal of content following a report | Decreased from 41% to 27% |
| Children (9–16) who say restrictions would limit their freedoms and rights | 45% |
Three months after Australia's under-16 media ban, the country's eSafety Commissioner found that around 70% of affected users had managed to keep their accounts on Instagram, Snapchat, and TikTok.
"There is no silver bullet to protect minors online. It takes a set of measures and actions that several stakeholders can collaborate to bring to life," says Giulia Grandin, policy analyst at the Lisbon Council.
Regulation: DSA and the upcoming Digital Fairness Act
Giulia Grandin singles out the Digital Services Act (DSA) as a "vital investigative tool" for boosting child safety online. The DSA places the burden of proving a service is safe onto the platform offering that service.
"The Commission announced another legislative tool, the Digital Fairness Act, complementing its DSA arm that would focus on commercial practices (such as dark patterns, marketing by influencers or design rather than content)," she explains.
"Social media is not a monolith. A private group chat for school is useful for communication and connection; an infinite scroll feed tuned for dopamine is a risk," says Grandin.
President von der Leyen's special panel on child safety recommends staged access by age, supervision for the youngest, and safe defaults for adolescents. Grandin emphasises that the most powerful safeguard is requiring that the moment a platform identifies a minor, its entire system must flip to work in that child's best interest.
"Safety should be the default setting... an online space that is safe for children is safer for all users, of all ages and genders," adds Grandin.
What young people are asking for
"Minors are not asking to be banned. They are asking for control, inclusion and participation," says Grandin.
Her recommendations from dialogue with youth representatives include:
- Hands-on digital literacy
- Platforms that are safe by design
- Investment in offline spaces
She also highlights the importance of online spaces for marginalised youth, including LGBTQI+ youth and those in isolated areas.
"True digital safety is not measured by how the average user navigates a platform, but by the impact the environment has on the most vulnerable," says Grandin.
From crisis response to prevention
Education in online safety is still aimed mostly at secondary schools, while first exposure happens years earlier. EIGE's recommendations include:
- Mandatory, gender-responsive digital literacy across primary and secondary education, covering digital identity, online relationships, misinformation, AI-generated content, and online safety
- Co-design of programmes with young people from the ideation phase to implementation
- Bystander intervention built into education
- Education aimed squarely at boys and young men, who make up the majority of abusers
"We often talk about young people, without them. There should be nothing about them, without them," says Giulia Grandin.
"Ending gender-based cyber violence means changing the environments in which young people grow up, both online and offline. That means safer platforms, better education and bringing boys and young men into the solution," adds Carlien Scheele.