A new study suggests algorithms shape user behaviour more powerfully than most people realise, adding weight to calls for stricter platform oversight in Denmark and across Europe.
The invisible hand guiding what you watch, click, and buy online may be more powerful than you think. According to DR, recent research indicates that recommendation algorithms on social media and video platforms measurably alter user behaviour, often without users being aware of it. The findings arrive as European regulators tighten the screws on big tech, demanding transparency about how these systems actually work.
Not Just Sorting, Steering
The debate has shifted. It is no longer whether algorithms sort content, but whether they actively steer attention, emotions, and decisions. These systems are designed to maximise engagement, watch time, or purchases. That means content triggering strong reactions or quick clicks gets promoted. Over time, users can be drawn into narrower streams of information. What feels like personal choice may be partly algorithmic design.
I have watched this play out in Denmark for years. We are heavy users of digital platforms, from news sites to TikTok. What algorithms choose to show us shapes not just entertainment, but politics and public debate. When a platform decides what gets prominence, it decides what feels normal or important.
How It Works
Recommendation systems influence through repetition, ranking, and personalisation. They do not invent content, but they decide what you see first. Material that keeps you scrolling or watching longer tends to win. Research shows measurable effects, though the size varies by platform and user group. Some studies find algorithms increase exposure to extreme or sensational content. Others show they help users find relevant material faster.
The truth is probably both. Algorithms can reduce information overload. They can also amplify outrage and distortion. The mechanisms are the same. The outcome depends on what the system optimises for. Right now, most platforms optimise for engagement. That is not a neutral goal.
Young Users at Risk
Children and teens may be especially vulnerable. Algorithms that reward compulsive use can disrupt sleep, encourage social comparison, and expose kids to unsafe content. This has become a major policy concern in Europe. Denmark, like other EU countries, is pushing for age appropriate design and default safety settings. Critics say platforms offer parental controls, but they are inconsistent and hard to use.
The evidence on harm is still debated. But the risk is significant enough to drive regulation. And for parents trying to manage screen time or explain why their kids cannot stop watching, the algorithms are not a side issue.
Europe Responds
The EU’s Digital Services Act is now the main legal framework. It requires very large platforms to disclose how recommendation systems work, conduct risk assessments, and offer users more control. The regulation treats algorithmic influence as a structural issue, not just a personal problem. Platforms must prove their systems do not unlawfully amplify harm or manipulate choices.
That is a shift. For years, big tech insisted algorithms simply respond to user preferences. Now regulators are saying prove it. The catch is that platform algorithms remain largely opaque. Companies argue that disclosure risks trade secrets. Researchers often cannot inspect the systems fully. So public debate outpaces hard evidence.
Denmark’s Stake
This matters here. Denmark’s media landscape is small and highly digitalised. Platform ranking choices affect which news stories reach people, how political debates unfold, and what feels culturally relevant. Danish authorities are increasingly focused on consumer protection and democratic integrity online. Algorithmic power is not just a Silicon Valley issue. It shapes daily life, from what Danes read to what they buy.
I have also seen how Danes engage with AI in other contexts, including health and even religious services. Trust in technology runs high here. That makes transparency even more important.
What Comes Next
Platforms say users have control through settings and feedback tools. Critics respond that those controls are often hidden or less prominent than default algorithmic feeds. Even when choice exists, defaults are powerful. And real transparency remains elusive. The most defensible claim is this: algorithms do not determine behaviour in a simple way, but they measurably shape what people see and think is important.
The policy environment is moving toward accountability, not away from it. Whether that will be enough is an open question. For now, the evidence suggests we are all being nudged more than we realise.
Sources and References
DR: Studie tyder på algoritmer gør mere ved os end du måske lige tror
The Danish Dream: AI skills now essential in Danish job market
The Danish Dream: Danes turn to AI like ChatGPT for diagnoses
The Danish Dream: Denmark debates AI sermon tool for pastors
