The German government is planning a law against digital violence to make it easier for those affected to enforce their rights in the digital sphere and to close gaps in protection. In your view, how serious is the problem at present, particularly with regard to hate speech?
Hate speech is a significant societal challenge. It is by no means a new phenomenon – insults, denigration and discrimination have always existed and are also encountered by people in the offline world. The key difference, however, is that such content spreads much more quickly in digital spaces, remains permanently visible and can potentially reach a very large audience.
Various studies from recent years demonstrate just how widespread the problem has become. According to these, at least one in two users has already encountered offensive or derogatory content on social media. For some, insults have become an integral part of their online experience.
And to what extent does hate speech influence debates in the digital sphere?
Digital spaces for debate are being permanently altered. For those affected, hateful comments can cause considerable psychological distress – such as stress, anxiety or the feeling of being pushed out of the public sphere. The social repercussions are also particularly problematic:
If users perceive digital spaces as hostile or unsafe, they are more likely to withdraw from discussions, express their opinions less frequently, or refrain from political participation altogether. As a result, important voices may disappear from democratic debates, whilst extreme or particularly vocal positions become disproportionately visible.
Steliyana Doseva To the profile
In recent years, hate speech has also become significantly more visible to the public. One factor contributing to this is that public figures – including, for example, politicians, journalists and researchers – are repeatedly targeted by hate and hostility online. The media attention such cases receive has raised awareness of hate speech as a societal problem and a form of digital violence. Furthermore, various regulatory initiatives at national and EU level have further highlighted the societal and political significance of the phenomenon. The intensified public and political debate shows that hate speech is no longer regarded as a marginal phenomenon of digital communication, but rather as a challenge with far-reaching consequences for public discourse.
Platforms are now to be held more strictly accountable by law. What responsibility do social media platforms bear in this regard?
Social media operators bear a key responsibility in dealing with hate speech, digital voyeurism, cyberstalking and other forms of digital violence. This is because platforms are no longer merely neutral technical infrastructures on which users publish their content. They are actively involved in curating content and, through the use of recommendation algorithms and moderation decisions, help determine what content people are shown.
This is particularly relevant because social media has now become an important source of news and information for many people. Younger users in particular regularly come into contact with news content via platforms such as Instagram, YouTube or TikTok. For Germany, the Reuters Institute Digital News Report 2026 shows that the importance of social media as a means of accessing news has continued to rise across all age groups: at 30 per cent, social media is already the second most common way of accessing online news. In the 18–24 age group, almost one in two people regularly use social media to access news.
How can social media platforms fulfil this responsibility?
If platforms are involved in opinion-forming processes, they must also take responsibility for how content that threatens democracy, is discriminatory or is offensive is handled. Consistent content moderation is crucial, particularly when dealing with offensive content. This includes transparent rules, easily accessible reporting channels and swift responses to illegal or harmful content. It is important that platforms also take on structural responsibility: for example, by systematically reviewing which content is made particularly visible through their recommendation systems and whether these mechanisms encourage hate, harassment or radicalisation.
At the same time, promoting respectful digital interaction is a shared responsibility. Users bear responsibility for their own communication behaviour. Media literacy and civil society ‘counterspeech’ remain key building blocks in tackling digital violence. At the same time, platforms shape the framework within which digital communication takes place. They therefore bear a special responsibility to promote safe spaces for debate and to effectively limit risks such as hate speech or harassment.
You also conduct research into the aforementioned ‘counterspeech’ – that is, public opposition to hate speech. Why is counterspeech considered an important approach to tackling hate speech?
Counterspeech offers a way to publicly challenge discriminatory or misanthropic statements and to highlight democratic standards of communication. Research findings show that certain forms of counterspeech can help to reduce the proportion of hateful content in discussions and promote more constructive forms of communication.
Steliyana Doseva To the profile
Which forms of counterspeech work particularly well in this context?
Forms of counterspeech that appeal to empathy and reason, or remind people of respectful standards of communication, are particularly effective. Messages of support for those affected can also be important, as they mitigate the negative consequences of hate speech and help prevent escalation. By contrast, purely punitive reactions are less successful. They often fail to bring about a lasting change in mindset and can sometimes even provoke defensive reactions.
At the same time, research shows that very few people actually intervene when they come across insults online. There are various reasons for this: some feel unsure or do not know how to react. Furthermore, more subtle forms of hate speech, in particular, are not always clearly recognised. The atmosphere of discussion on platforms also plays a role: in hostile environments where counter-speech receives little support, users are often less willing to step in – partly because responding to insults, for example through counter-speech, takes time and energy.
How could regulation, platform moderation and counterspeech work together effectively?
Regulation, platform moderation and counterspeech should be understood as complementary approaches. To effectively curb digital violence and hate speech, legislators, platform operators and users all bear a share of the responsibility.
An important task for legislators is to establish a clear legal framework. This includes, in particular, a definition of hate speech that is as consistent and transparent as possible. On this basis, platforms can implement their moderation measures, for example by identifying and removing unlawful content. At the same time, they should ensure transparency regarding how decisions are reached.
By comparison, counterspeech fulfils a completely different function: it neither replaces regulation nor moderation, but complements them. Furthermore, counterspeech enables societal responses to problematic content that is not necessarily unlawful. It is precisely in such grey areas that public opposition from users can be an important tool.
To what extent can AI provide support here in future – and what new challenges lie ahead?
AI-supported systems could play an important role in the future. They can help people formulate appropriate and constructive responses to hate speech more quickly. Furthermore, AI offers potential for individual and respectful dialogues over a longer period of time. Such conversations would be virtually impossible for human counterspeakers to carry out on a large scale, partly because they are very time-consuming.
To ensure that such AI-supported systems can be used responsibly, clear legal guidelines and transparency are needed regarding how they work and what data they use. At the same time, researchers and regulators need access to suitable data in order to assess the effectiveness and potential risks of such systems.
It is also important to give users greater control and a better understanding of how their data is used and how platforms use AI to make decisions about their content. At the same time, AI development currently finds itself caught between the need for large volumes of data, the protection of data privacy and copyright, and legitimate economic interests. The challenge for regulation therefore lies in facilitating innovation and access to data without losing sight of the rights of those affected.
What would you like to see from policymakers and platforms to make digital communication spaces safer and more constructive in the long term?
Above all, I would like to see policymakers, platforms, civil society and the research community working more closely together to better understand which measures are actually effective and what unintended consequences they have. Digital communication spaces are changing very rapidly. That is why we need more evidence-based decisions rather than short-term reactions to individual debates or new technologies.
In my view, this also includes greater transparency. Platforms should provide researchers, regulatory authorities and civil society actors with greater insight so that the effectiveness of moderation measures and new AI systems can be independently assessed. At the same time, users should be better able to understand what data is being used and how decisions regarding their content are reached.
Sources
Survey-Summary in English available
bidt-publications
