On TikTok or Instagram, what you see in your advertising feed depends much more on your socio-economic profile than on your tastes. Behind every video touting easy income, miracle training or a winning bet, there is an algorithm that has gotten to know you, often better than you imagine. While some see this as progress in terms of personalization, others denounce a new form of invisible discrimination, which primarily affects the most vulnerable young people.
How Platforms Guess Your Social Status Without Ever Asking
Social networks don't need to ask you direct questions to understand your reality. By analyzing discrete signals (the type of phone used, the IP address, the accounts followed or the content viewed), their artificial intelligence systems infer your income level, your education or your professional situation. This ability to deduce is perfected as you interact, creating a kind of increasingly precise digital double.
It is on this basis that sponsored content is distributed. A study carried out by Carolina Sáez-Linero and Mònika Jiménez-Morales for the University of Navarra showed that cross-referencing the postal addresses of young people surveyed with the Catalan Territorial Socioeconomic Index (TSI) made it possible to establish striking correlations between real profiles and the advertisements received. Algorithms do not need to read a pay slip to know who is likely to be seduced by a promise of immediate credit.
Among the 1,200 young people aged 14 to 30 surveyed, those from modest backgrounds reported greater exposure to certain advertisements. This content highlighted quick and easy gains, such as influencer training or cryptocurrency investments. They also praised profitable teleworking without experience or even loans obtained immediately. This selection of content is not trivial when it occurs on screens that adolescents consult for several hours a day.
Advertising algorithms exploit economic vulnerability online
These systems do more than reflect personal preferences. Little by little, they learn to get ahead of them, then to strengthen them. They may even end up shaping them. In a constant flow of sponsored content, certain messages end up taking over. Thus, in an adolescent, the promise of rapid social success can create an algorithmic belief. He comes to think that his future follows the path that the platforms trace for him.
The study, relayed by Phys.org, confirms that these contents do not appear randomly. The most disadvantaged young people are twice as likely to be exposed to advertisements promoting miracle financial solutions as their counterparts from more affluent backgrounds. One of the most notable cases remains that of online betting and games, the frequency of which increases markedly with insecurity.
This logic does not always result from a voluntary choice of brands. Often, the algorithm optimizes efficiency by exposing the most vulnerable to the most profitable content. However, they are the ones who click or convert the most. Researchers show that even neutral campaigns can drift toward social bias when automatic optimization gets involved.
In this context, the algorithm acts as a commercial actor in its own right. It exploits, without giving the impression, the economic fragilities, the desires for autonomy and the dreams of the youngest. At 17, a teenager does not always perceive the persuasive nature of what he is watching. This is even more true when the message slips into a viral video, carried by a familiar tone. In addition, minors have difficulty locating advertising content. This becomes complex when they are mixed with native formats, carried by influencers or hidden in trends.
Why do these mechanisms hit precarious young boys harder?
Platforms do not apply targeting logic uniformly to all profiles. The study shows that advertisers particularly target young boys from modest backgrounds with advertisements for gambling, cryptocurrencies or sports betting. On the other hand, algorithms much more often expose young girls, even at the same socio-economic level, to beauty, parenting or lifestyle content.
This difference is partly explained by the way the algorithms cross-reference gender and class variables. An 18-year-old boy living in a working-class neighborhood will therefore see an advertisement for a trading application more often than a boy of the same age living in a wealthy neighborhood. As for a girl of the same age and from the same neighborhood, she will not receive these same incentives to earn money, but rather suggestions for skincare products, trendy clothes or content related to education.
This double filter thus creates very different exposure trajectories. For boys in precarious situations, it creates a digital environment saturated with messages about quick money, risk and competition. This climate can encourage impulsive decision-making, especially among those who feel excluded from traditional paths to success. The repetition of these messages acts as silent conditioning, which guides aspirations and behaviors, sometimes lastingly.
Far from being a simple side effect, this algorithmic stratification fuels inequalities from adolescence onwards. It produces a world where everyone receives content shaped not only by their tastes, but also by the machines' perception of their social value. As long as these mechanisms remain opaque, they will continue to reproduce (or even aggravate) the gaps they claim to ignore.
With an unwavering passion for local news, Christopher leads our editorial team with integrity and dedication. With over 20 years’ experience, he is the backbone of Wouldsayso, ensuring that we stay true to our mission to inform.
