EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

Europaudvalget 2024-25
EUU Alm.del Bilag 554
Offentligt

Case no.

2025 - 2150

Document no.

143397

Date

13/06/2025

The Danish government’s response to the Commissions consultation re-

garding Article 28 of the Digital Services Act

The Danish Government would like to express its appreciation for the chance to respond to

this consultation. Protecting minors online is a key priority of the Danish government to which

strong enforcement and effective implementation of the DSA is pivotal. Article 28(1) is the

heart of the DSA’s approach to the protection of minors on online platforms, and accordingly,

the guidelines will be central in ensuring the effective enforcement of article 28 in all Member

States.

The Danish Government

welcomes the Commission’s draft guidelines,

which hold great po-

tential to improve the privacy, safety and security for minors online. However, in order for the

guidelines to have the desired effect and realize their potential, it is crucial that online plat-

forms and especially social media correctly identify and categorize their users as being minors

or not. Unfortunately, the draft guidelines fall short of this objective.

The draft guidelines’ proposed approach of relying primarily on

age

estimation

for major so-

cial media platforms is insufficient given the comprehensive risks these platforms can pose to

minors and given their documented ineptitude at keeping minors from their service using

other age estimation techniques. The Danish Government finds it to be of utmost importance

that the guidelines are amended to require availability of robust age

verification

mechanisms

as the default standard for all platforms, presenting a wider range of risk categories to mi-

nors.

In the following, we elaborate our key messages regarding the draft guidelines:

•

Clear recommendations on effective and privacy-preserving age verification;

A broad scope which covers all platforms that are actually accessible to minors;

Welcoming the risk-based approach, but social media platforms should be categorized

as high risk per default;

Ensure that minors are not exposed to retention mechanisms and addictive designs

and make use of default settings to ensure a baseline of safety for minors;

AI Chatbots should be disabled by default, not require payment to opt out of and be

free of manipulative designs;

Making it as easy to report illegal and harmful

content as it currently is to “like” or

“share” content.

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

Clear recommendations on effective and privacy-preserving age verification

The guidelines should make it crystal clear that all relevant online platforms and especially so-

cial media should implement age verification to comply with Article 28 of the DSA.

If providers are to ensure safe and secure platforms with age-appropriate content and inter-

faces, the provider must have a firm awareness of the age and maturity level of their users.

Effective age verification is a prerequisite for this. That is made abundantly clear, as the major

social media platforms use age estimation techniques, but still fail to keep children below

their own age limits from their services. A study from 2024 shows that 48 percent of all Dan-

ish children have a profile on social media before they turn 10 years old, whereas 94 percent

have a social media profile before the age of 13. This is despite the fact that major social me-

dia platforms by their own accounts use ‘effective’ age estimation and typically do not allow

children below the age of 13 on their platforms.

The draft guidelines’ proposed approach of relying primarily on age estimation for major so-

cial media platforms is insufficient given the comprehensive risks these platforms can pose to

minors. Many online services and platforms can be harmful for minors if accessed without

guardrails and protections. Therefore, efficient age verification tools are needed in order to

protect minors in digital environments. Such tools are under rapid development, and will

soon be available across many EU member states.

Age estimation techniques have significant accuracy limitations, with error rates that may be

hard to prove or investigate, and with a lack of transparency in terms of their application. Ad-

ditionally, such estimation techniques may be opaque and very data-intensive in terms of

analysis of user behavior, and may sometimes be bordering on profiling of users.

Age verification may introduce a small friction for users. However, it is a proportionate meas-

ure to protect minors from the risks (content, conduct, contact, consumer, cross-sector) on

many online platforms. In the offline world, age checks are standard for age-restricted goods

and services. Thus, it is perfectly reasonable to expect similar safeguards online, where the

risks

—

especially for children and teens

—

are significant and well-documented.

Against this background, the Danish Government strongly recommends that the draft guide-

lines are amended to require availability of robust age verification mechanisms as the default

standard for all platforms, presenting a wider range of risk categories to minors. Specifically:

•

Online platforms and especially social media should be required to enable and imple-

ment age verification relying on the technical specifications used in the forthcoming

EU age verification app and EUDI Wallet.

The guidelines should establish clear criteria for when alternative approaches may be

permitted, with the burden of proof resting on the platforms to demonstrate the effi-

cacy of less stringent measures.

•

The guidelines envision an EU age verification app primarily targeted at 18+ restricted content

like pornography and gambling. The Danish Government suggests, that the guidelines make it

clear, that where more granular age-verification is available, all relevant online platforms and

especially social media should make use of these.

The Danish Government believes that the upcoming EU age verification app and the European

Digital Identity (EUDI) Wallet will provide a commonly available, seamless and privacy-pre-

serving way to verify age, which will minimize friction especially if the verification method is

broadly adopted, and thus familiar to the user. While some member states may not be able to

issue +13 or +15 age attestations, many member states are in fact able to do so, or will be

able to do so in the future. Protecting minors online using age verification in these member

states should not be limited by the fact that other member states do not have such technical

solutions ready. Especially since it will not substantially increase the technical complexity for

service providers to utilize the age verification solutions.

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

Please find our proposed concrete changes to the draft guidelines regarding age verification

in the annex.

A broad scope which covers all platforms that are actually accessible to minors

In order to ensure the same level of protection for all minors in the EU, it is essential that all

online platforms where children are present, adhere to the guidelines. Hence, we appreciate

how the Commission has sought to further clarify which online platforms are obligated to

comply with Article 28. We welcome that it is expressed clearly that also online platforms

which state in the terms and conditions that the platform is restricted to users above 18 years

of age, shall comply with the guidelines where users under the age of 18 in fact use their ser-

vices.

However, the interpretation of how providers can become aware of the fact that some recipi-

ents on their services are minors, could be improved. As currently phrased, the examples are

limited to situations where the provider processes relevant personal data or has conducted or

commissioned its own research on whether minors access their platforms. This does not con-

sider information from external sources, including academic studies, civil society organiza-

tions and regulatory bodies, which often document widespread use of platforms by minors. In

several instances, this type of evidence has been publicly available without being acknowl-

edged or acted upon by platform providers. We therefore recommend that the guidelines

clarify that a platform is also considered to be aware of minors accessing its services, where

independent and credible third-party evidence points to the presence of minors on that plat-

form.

Following a risk-based approach to the protection of minors, but social media platforms

should be categorized as high risk per default

The Danish Government welcomes how the draft guidelines adopt a risk-based approach, rec-

ognizing that different platforms pose varying levels of risk to minors. Accordingly, the draft

guidelines recommend that each platform tailor their measures to their specific services.

While we agree with this approach, it should be redundantly clear, that all social media plat-

forms covered by Article 28 of the DSA are considered a high-risk category and that they must

adhere to the highest standards set out in these guidelines.

Social media platforms pose significant risks to minors. They entail a juxtaposition of functions

posing a variety of risks to their users, including exposure to harmful content, addictive de-

signs, commercial profiling, and retention mechanisms, harmful encounters with adult preda-

tors, and many more. As a result, European children are too often exposed to illegal or harm-

ful content and comments, and faced with unwanted contacts online. Social media algorithms

push extreme content to vulnerable minors, which can cause or exacerbate mental health

problems, including self-harm, poor body image or eating disorders. Recent years have seen a

sharp rise in problematic social media use among minors in Europe, spurred by retention

mechanisms intentionally designed to catch and keep minors’ attention. Such problematic so-

cial media use can cause addiction-like symptoms, lower mental and social well-being and is

even correlated with higher substance use. The guidelines must ensure that social media plat-

forms understand that they carry the responsibility to reverse this development.

In addition, a risk-based approach should take into account that many online platforms have

mixed target groups and users. This entails outlining clear and practical definitions of what

constitutes inappropriate content for minors and how the particular type of content should

be categorized based on its harmful nature, in order to help the individual industries to iden-

tify and mitigate risks.

The annex included with the draft guidelines, presenting the OECD 5C typology of risks, are

hence not only useful, but necessary. While we can support the use of the OECD typology, we

nevertheless find that it would prove useful for implementation and enforcement purposes to

include further examples of the different risk categories. Most pressingly, the Danish Govern-

ment would urge the Commission to include content promoting “self-harm” among the exam-

ples of “Content risks”. Studies show that more than

21 percent of Danish children in the

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

ninth grade have committed self-harm. The sharing and promotion of self-harm content on

social media is a recurring problem, that needs to be addressed.

Ensure that minors are not exposed to retention mechanisms and addictive designs and

make use of default settings to ensure a baseline of safety for minors

Default settings can play a crucial role in protecting minors online by offering a baseline of

safety without requiring a proactive effort from the minor. The Danish Government hence

welcomes how the draft guidelines recommend implementing default settings which create a

consistent layer of protection. We further support how the draft guidelines allow for parental

control mechanisms, but at the same time recommends default settings ensuring that even

less tech-savvy families benefit from safety measures, reducing the chances of minors being

accidentally exposed to risks.

We also appreciate that the Commission has responded to our call for an exhaustive descrip-

tion of useful and recommended default settings, including time notifications, turning off "ge-

olocation", "seen" and "read" function and the recommendation to make some functions less

visible, e.g. the "like" function.

Of particular importance is to ensure that minors are never exposed to retention mechanisms

and addictive designs when engaging on online platforms. For this reason, it is central to

maintain the recommendation on online interface design in the draft guidelines, which state

that online platforms should ensure

“that minors are not exposed to persuasive design fea-

tures that are aimed predominately at engagement or that may lead to extensive use or over-

use of the platform

or the forming of problematic or compulsive behavioural habits”,

as well

as maintain the useful examples of such design features, including

“infinite scroll”

and

“auto

play”.

AI Chatbots should be disabled by default, not require payment to opt out of and be free of

manipulative design

The Danish Government appreciates that the draft guidelines acknowledge the specific risks

posed by AI systems such as chatbots and filters, including the need to clearly inform minors

when they are interacting with non-human agents. However, transparency is not enough to

protect children from the risks posed by AI-chatbots,

and it should not be the children’s re-

sponsibility to ensure that online platforms are safe environments for them to engage in.

Certain AI-chatbots have been shown to emotionally manipulate with children and create a

false sense of safety and relational closeness. They have further ended chats with suggestions

for the child to

“share

more”,

noting that the chatbot is “there for them”. This kind of emo-

tional engagement retains children on the platform and increase their dependency of it.

We therefore urge that the guidelines recommend that AI-chatbots are disabled by default

and that they should be free of manipulative design. Minors should not be locked into interac-

tions with a chatbot which can influence their behavioural, buying and spending patterns.

Some online platforms require that users pay for removing the AI-chatbot once installed, why

it should be made clear that such practice is unacceptable. Minors should always be able to

remove AI-chatbots free of charge.

We further recommend that the guidelines emphasize that AI chatbots must be designed and

deployed in ways that are demonstrably safe for children. This includes ensuring that AI-chat-

bots are not only labelled, but that they do not provide advice or guidance on sensitive topics

without appropriate safeguards, validation and clear limitations. AI systems can inadvertently

offer misleading or developmentally inappropriate suggestions, particularly in response to

children’s queries related to mental health, relationships, or body image and should therefore

be subject to strict content boundaries and human oversight.

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

Making it as easy to report illegal and harmful

content as it currently is to “like” or “share”

content

The Danish Government welcomes the Commission’s focus on

reporting and namely on how

the online platforms should make it easy for minors to report illegal, harmful or other problem-

atic content. The sharing and promotion of illegal or harmful content remains a core problem

of especially social media platforms.

A Danish study has found an increase in all the different kinds of unpleasant experiences mi-

nors can encounter online in the years between 2021 and 2024. Accordingly, 32 pct. of Danish

children have experienced violent photos or videos, they did not want to see. 10 pct. have ex-

perienced threats or blackmailing, which is double the number reported in 2021. 9 pct. have

seen their private or intimate photos shared against their will, against 3 pct. in 2021.

Despite of an increasing focus on the need to protect minors online and in spite of the fact

that the DSA has entered into force in this time period, the changes we see, are not for the

better. In Denmark, the experience is that minors seldomly utilise the option of reporting ille-

gal content to the online platforms, as they either are not aware that this option exists or as

they find it difficult to report the content in practice. In order to ensure that the reporting

mechanisms put in place by the DSA have an actual effect when it comes to protecting mi-

nors, we find that the guidelines should recommend that online platforms make it as easy to

report content as it currently is to “like” or “share” content.

Further, special consideration should be given to the online platforms’ treatment of feedback

received from minors. The text should be strengthened to ensure that feedback from minors

is not only accommodated, but actively prioritized. Both with a view to enable swift take-

down of harmful content spreading through sites populated by minors, but also with a view to

learn from minors’ perceptions of harmful content. Minors’

experiences, needs and percep-

tions of harm often differ significantly from those of adults and their feedback may reveal dis-

tinct risks related to exposure to harmful content. We therefore urge the Commission to rec-

ommend that providers of online platforms are required to collect and analyze

children’s

feedback separately from that of adults as a distinct category. The resulting findings should be

directly reflected in platform decision-making processes concerning moderation and in adap-

tations of recommender system and interface design.

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

Annex: Specific Comments

PROPOSED CHANGES (non-exhaustive, but provided to give an indication);

Current wording

Ref.

(from

line)

231

Proposed new

text

Justification

Before deciding whether to

put in place any age assur-

ance method, providers of

online platforms accessible

to minors should always

conduct an assessment to

determine whether such a

method is appropriate to en-

sure a high level of privacy,

safety and security for mi-

nors on their service and

whether it is proportionate,

or whether such a high level

may be achieved already by

relying on other less far-

reaching measures

Providers of

online platforms

accessible to mi-

nors should imple-

ment appropriate

age verification

measures as a

standard protec-

tion mechanism

–

preferably based

on the technical

specifications and

standards used by

the EU Age Verifi-

cation App and

the EUDI Wallets.

Alternative ap-

proaches may be

implemented

when the provider

can demonstrate

with clear evi-

dence that such

alternatives de-

liver equivalent or

superior protec-

tion for minors.

DELETE

The current wording is prob-

lematic because it:

Positions age verifica-

tion as an exceptional

measure requiring spe-

cial justification

Characterizes age as-

surance as potentially

disproportionate or

"far-reaching"

Creates a presumption

that alternatives should

be preferred

The proposed amendment

properly recognizes age veri-

fication as a fundamental

child protection tool rather

than a burden requiring spe-

cial justification. It aligns with

our established understand-

ing of comprehensive online

risks to minors while ensur-

ing proportionate implemen-

tation tailored to each ser-

vice.

We find it problematic that it

is the provider of the online

platform which solely evalu-

ates and assesses the risk.

This enables platforms to pri-

oritize business interests

over minors’ safety.

In case the above line 231 is

amended, this part can be

deleted.

259

Any other circumstances in

which the provider of an

online platform accessible to

minors has identified high

risks to minors’ privacy,

safety or security, including

contact risks as well as con-

tent risks, that cannot be

mitigated by other less intru-

sive measures

284

Where the provider of the

online platform has identi-

fied at least medium risks to

minors on their platform as

established in its risk review

(see Section 5 on Risk Re-

view) and those risks cannot

be mitigated by less restric-

tive measures.

DELETE

Same as above. Additionally,

the guidelines appear to ac-

commodate existing techno-

logical constraints by requir-

ing less rigorous age assur-

ance measures for platforms

with age thresholds below

18, despite these platforms

potentially posing significant

risks to younger minors. Age

EUU, Alm.del - 2024-25 - Bilag 554: Notat og høringssvar vedr. Kommissionens høring om beskyttelse af børn online

verification should also be

usable for these services.

290

Providers of online platforms

accessible to minors that are

confronted with those two

scenarios may also opt to

put in place age verification

methods instead. In any

event, providers should con-

duct a proportionality as-

sessment justifying the

adoption of age assurance

measures prior to putting

them in place.

Providers of

online platforms

accessible to mi-

nors that are con-

fronted with those

two scenarios are

encouraged to put

in place age verifi-

cation methods to

complement any

age estimation

measures. In any

event, providers

should conduct a

proportionality as-

sessment justify-

ing the adoption

of age assurance

measures prior to

putting them in

place.

EU age verification

solution, including

an app, will be an

easy-to-use age

verification

method that can

be used to prove

that a user is 18 or

older (18+) and in

many Member

States also other

relevant age

groups.

The Commission

will review these

guidelines as soon

as this is neces-

sary in view of

practical experi-

ence gained in the

application of that

provision and the

pace of technolog-

ical, societal, and

regulatory devel-

opments in this

area, and in no

later than 5 years.

Privacy preserving age verifi-

cation should always be en-

couraged where possible.

The current wording seems

to hint that it is an extraordi-

nary measure not to be used

unless necessary. We disa-

gree with this approach (see

also counterarguments sec-

tion above)

EU age verification solution,

including an app, will be an

easy-to-use age verification

method that can be used to

prove that a user is 18 or

older (18+).

Adjusting the wording is nec-

essary to reflect that +18 is

not the only age possible to

verify

–

for many MS it

should be possible to obtain

and issue age attestations of

other age groups (+13) or

(+15) for example.

1020

The Commission will review

these guidelines as soon as

this is necessary in view of

practical experience gained

in the application of that

provision and the pace of

technological, societal, and

regulatory developments in

this area.

Timely review of the guide-

lines will be necessary in or-

der to evaluate their effec-

tiveness, incl. their propor-

tionality. As stated, the area

is quickly developing. The

general evaluation of the

DSA can also lead to updates,

which will need to be re-

flected in the guidelines.

This proposal does not

preempt the Commission to

review the guidelines sooner

than 5 years, should they

deem it relevant.