
How Social Media Platforms Weaponize Approval to Capture Adolescent Attention
A thousand likes, and no one is watching.
Social media's core business model is the artificial inflation of social approval signals, creating a structural race to the bottom that exploits adolescent neurobiology in ways categorically different from ordinary developmental adversity — and no amount of internal wellbeing optimization can fix what is architecturally broken.
Actions
The Observer
The Translation
AI-assisted summaryFamiliar terms
The central argument here reframes the social media harm debate by identifying the social feedback mechanic not as a feature but as the foundational business architecture. TikTok's competitive strategy against Instagram illustrates this precisely: by inflating reward signals — obscuring whether displayed numbers represent likes, views, or unique users — the platform injected what functions as a social approval growth hormone. The numbers are often physically impossible as representations of real engagement, but they manufacture the sensation of a vast awaiting audience. Because platforms compete for attention on a common axis of social validation potency, a structural race to the bottom emerges. Whoever delivers the most concentrated hit of approval captures the user's time budget.
The exploitation extends beyond consumption into labor extraction. TikTok's Discover tab presents branded hashtag challenges with billion-scale view counts, creating the illusion of audience demand. Adolescents then generate advertising content — unpaid — driven entirely by the promise of social feedback. Platform-employed wellbeing teams, however credentialed in positive psychology or subjective wellbeing research, face an intractable problem: the product's core mechanic is precisely what they would need to dismantle.
The insight draws a categorical distinction between ordinary developmental adversity and algorithmically mediated social evaluation. Antifragility theory holds that stressors build resilience when they are proportionate, recoverable, and embedded in contexts permitting agency. The social media environment facing a twelve-year-old girl meets none of these criteria: it is chronic, ambient, socially inescapable, and amplified by recommendation algorithms that make social status numerically legible and publicly ranked among real peers. The same intellectual framework that warns against overprotecting children — Haidt and Lukianoff's antifragility argument — becomes the framework that explains why this specific environment is not building resilience but systematically eroding it.
