What a fact-checking chatbot teaches us about young people's attention
We just launched the pilot test of WOBSONGO, a WhatsApp chatbot designed to help young people in Burkina Faso check sexual and reproductive health information before believing it or sharing it. The initial cohort brought together volunteer youth testers recruited from different provinces. Within the first few weeks, two numbers stopped us in our tracks.
<30% actually tested the tool. Of all the young people included in the cohort.
This is valuable data is telling us something important about young people's relationship with information, something that goes well beyond our tool and deserves to be discussed openly.
1. The real problem isn't only access to information
One could attribute this under-use to technical causes: unstable connections, the cost of data, messages not getting through. That's a real avenue, one we are actively investigating with the young people themselves. But the technical hypothesis doesn't explain everything.
The real issue is behavioral: checking information is not a spontaneously felt need. No one wakes up in the morning wanting to check a rumor. On the other hand, many young people wake up wanting to know what's circulating, what's buzzing, what confirms what they already thought.
This is where false information structurally wins: it doesn't need to be true to be consumed. It just needs to be attractive.
2. Fake content isn't amateur, it's designed to win attention
On social media, information spreads because it delivers something concrete. The content that spreads best activates one or more of these levers:
• A strong emotion : fear, outrage, surprise
• A confirmation of belief : what we already thought, finally “proven”
• A sense of belonging : being part of the group of those who “know”
A traditional remedy that “works every time,” a theory that simply explains a complex phenomenon, a rumor that flatters an intuition already there: all of this is optimized to capture attention. Real information, on the other hand, is optimized for accuracy. It is often more nuanced, less spectacular, sometimes less immediate. It demands a cognitive effort that fake news spares you.
What the research says
A landmark MIT study (Vosoughi, Roy, and Aral, Science, 2018), covering 126,000 rumors and more than 4.5 million tweets, found that false information is 70% more likely to be shared than true information, and reaches around 1,500 people about six times faster. The reason isn't naivety: false content is more novel, more surprising, and triggers fear or disgust, emotions that drive sharing, whereas true content tends to trigger trust or sadness, emotions that are less likely to prompt a click.
The illusory truth effect. Fake content also wins through repetition. This is the effect documented by Fazio and colleagues (2015): the more often you hear the same piece of false information, the more you end up believing it's true, even when you knew from the start that it was false. More recent studies confirm that neither intelligence nor an analytical mind protects against this bias. This is not a problem of individual education: it is a universal cognitive mechanism, one that disinformation exploits by repeating itself until it takes lasting hold.
An organized industry. In West Africa, this mechanism is anything but amateur. According to tracking by the Africa Center for Strategic Studies, 189 disinformation campaigns have been documented across the continent, an increase of nearly 400% since 2022. West Africa alone accounts for nearly 40% of these campaigns (72 campaigns across 13 countries), with the Sahel particularly targeted since 2018. These campaigns rely heavily on WhatsApp and Telegram, exactly the channels young people use every day, and precisely the channel on which WOBSONGO chose to operate.
So it isn't that young people “believe anything” for lack of critical thinking. It's that two logics are clashing, attractiveness versus accuracy, and that the second one starts at a structural disadvantage, facing an opponent that has speed, emotion, repetition, and, increasingly, the resources of an organized industry on its side.
3. Informational agency cannot be decreed
We often talk about “raising awareness” or “strengthening young people's critical thinking” as if informational agency were a matter of moral willpower. The WOBSONGO experience shows something else: even a free tool, designed with young people, for their concrete health concerns, is not enough to generate the expected use.
Agency cannot be decreed. It is built through habit, through trust in the source, and above all through the pleasure of checking, not just the moral obligation to do so.
A fact-checking tool that feels like homework will always lose out to content that feels like a discovery.
4. What this teaches us, as solution designers
The under-use of WOBSONGO invalidates a widely held assumption: that easier access to accurate information is enough to change behavior. It's a useful lesson in humility for everyone designing digital solutions for young people in West Africa, in health as elsewhere.
The question is no longer just “How do we make accurate information accessible?” It becomes:
How do we make accurate information as desirable as false information? It's a paradigm shift: from disseminating information to designing experiences that make people want to check.
This is WOBSONGO's next challenge, and probably that of any initiative that genuinely wants to equip young people's informational agency, rather than simply giving them the theoretical means to do so.
WOBSONGO is a WhatsApp chatbot developed in partnership with KAIROS and ImpactScope, currently in a pilot phase in Burkina Faso. Its goal: to help young people in French-speaking West Africa check the health information circulating around them, before believing it, and before sharing it.
