EU Proposes Social Media and AI Chatbot Ban for Under-15s in Sweeping Child Safety Plan

EU Unveils Sweeping Social Media Restrictions and AI Chatbot Limits for Minors

The European Commission will propose the EU Kids Act on Thursday, introducing the most comprehensive child safety online regulations to date. The legislative package would restrict children under 15 from accessing social media, AI chatbots, video-sharing platforms and online games without parental approval, according to a confidential document obtained by Reuters and Euronews1

1

2

2

. Commission President Ursula von der Leyen is expected to preview the proposal during her State of the Union address to the European Parliament on Wednesday, with the full bill published the following day1

1

.

The ban on social media for under-15s targets major platforms including Meta's Facebook and Instagram, TikTok, Google's YouTube, and extends to AI-powered services like ChatGPT1

1

. This marks a significant expansion of child safety online efforts, combining two regulatory approaches that have developed separately: restricting social media access and limiting exposure to generative AI3

3

.

Graduated Age-Based Access System Balances Protection with Digital Benefits

The European Commission proposal implements a graduated approach to digital access based on age groups. Children under three would be barred entirely from the covered services2

2

. Those aged 3 to 12 can access only child-friendly services under full parental control with strict safety standards1

1

. Children between 13 and 14 could request parents to open introductory accounts for social media and video-sharing platforms, but these would operate under parental control with limited contacts and strict time limits1

1

.

From age 15 onwards, minors can set up their own accounts, though safety-by-design rules remain in place for teenagers aged 15 to 182

2

. "The Union's approach should allow minors to benefit from the huge potential of digital services, while protecting them from abuse and exploitation. It needs to limit the access of tech companies to our children, and not the other way around," the Commission document stated1

1

.

Tech Companies Face Mandatory Age Verification and Safety-By-Design Requirements

The proposal places substantial obligations on technology companies. Social media and video-sharing platforms must verify children's age when users open new accounts, while online games platforms must conduct age checks before allowing game downloads1

1

. Companies can use an EU-wide age-verification app or other national solutions that check only whether the user exceeds a certain age threshold2

2

. Proportionate verification is required for existing accounts, based on factors such as subscription duration2

2

.

Platforms must provide effective parental controls and tools enabling children to easily report harmful content1

1

. The safety-by-design approach requires companies to avoid addictive design features and harmful feeds1

1

. Specifically, the bill bans design features including infinite scrolling, artificial notifications and certain reward mechanisms2

2

. Recommender systems must be designed to prevent "rabbit holes" and avoid recommendations driven purely by engagement-based signals2

2

.

Enforcement Strategy Shifts Compliance Burden to Platforms Through Supervisory Fees

Companies will pay a supervisory fee to fund regulators' enforcement and supervision activities1

1

3

3

. The Commission is proposing to shift the compliance burden onto tech companies, requiring them to prove they are following child-safety rules2

2

. This addresses criticism that Brussels has been too soft in enforcing digital rules, as online environments tend to move faster than public authorities can monitor2

2

.

Minors' account content must be set to private by default, with unknown users unable to contact them2

2

. Risky settings should be disabled automatically2

2

. Online services already subject to the Digital Services Act's stricter regime must seek authorization from regulators2

2

.

Proposal Builds on Growing EU Regulatory Push Against Addictive Design

The EU Kids Act represents the culmination of a broader policy push that began with concerns about addictive design on social media. In November 2025, the European Parliament called for parental authorisation for children under 16 to access social media and asked the Commission to consider a harmonised digital age limit of 133

3

. By May 2026, Ursula von der Leyen announced the proposed Digital Fairness Act targeting features such as infinite scrolling, autoplay and push notifications3

3

.

In February, the Commission found that TikTok's addictive design could breach the Digital Services Act, marking the EU's first enforcement action focused on harmful platform design3

3

. By July, an expert panel recommended preventing children under 13 from accessing social media and other services with addictive or age-inappropriate features, while allowing stricter national limits3

3

.

Member States Push for EU-Wide Framework to Prevent Regulatory Fragmentation

France, Greece, Austria, Denmark, Spain, Belgium, Italy, Germany, Poland and the Netherlands are all considering or advancing national legislation on social media bans, creating pressure on Brussels to present an EU-wide measure and avoid fragmentation2

2

. The scope proved broader than originally anticipated, with the bill covering video games and AI companions following requests from Spain and the Netherlands2

2

.

The draft EU Kids Act specifically covers AI-powered chatbots and companions, described as "virtual tools that can give mental health and personal development advice to minors"2

2

. This puts services like ChatGPT in the same child safety online debate as social media platforms3

3

. Services designed for education or operated by public authorities, as well as AI systems intended for office or industrial use, are excluded from the scope2

2

.

International Context Shows Growing Global Movement on Child Digital Safety

Europe's debate follows a broader international push to limit children's access to social media and generative AI. Australia enacted one of the first national social-media age restrictions, while several governments including the UK have considered similar measures3

3

. Within Europe, France, Greece, Denmark, Sweden and Norway have pursued or considered their own age restrictions3

3

.

Norway has moved to sharply restrict generative AI use in schools, including a general ban for students aged 6 to 133

3

. New York City imposed a one-year prohibition on student-facing generative AI from 2-K through grade 83

3

. The Commission will need to negotiate the draft with EU member states and the European Parliament in coming months before it becomes law1

1

. Details may still change ahead of the announcement, and the final scope, exemptions and enforcement rules will depend on the legislative process3

3

.