How the Full Cost of Attention Theft Would Bankrupt the Companies That Profit from It

To understand what the full cost of attention theft would mean if internalized, it helps to understand what attention actually is and why it is worth stealing.

Attention is the gateway to cognition. Nothing enters understanding without first passing through attention. The quality of a civilization's thinking — its capacity to reason clearly about complex problems, to deliberate honestly about collective decisions, to sustain the kind of focused engagement that produces scientific discovery, artistic creation, or political wisdom — depends directly on the quality of attention that its members can bring to bear. Attention is not a luxury. It is the substrate on which everything else rests.

The attention economy does not steal attention in the dramatic sense. It does something subtler and more damaging: it systematically trains attention away from depth and toward reactivity. The design objective of engagement-maximizing algorithms is not to give users what they value but to trigger the neurological responses — novelty, outrage, social comparison, intermittent reward — that keep eyes on screens longest. These responses are real; they evolved for reasons that had nothing to do with algorithmic exploitation. But their exploitation at scale produces a population that becomes progressively less capable of the kind of focused, patient, effortful thinking that complex problems require.

The cognitive costs are documentable. Research on media multitasking shows that heavy social media users perform worse on tasks requiring sustained attention and cognitive control even when the devices are absent. The brain is not simply distracted in the moment; its operating mode is altered. Research on adolescent mental health shows alarming correlations with social media adoption timelines. The sudden, simultaneous rise in depression, anxiety, and self-harm across multiple countries in the early 2010s — precisely when social media use became pervasive among teenagers — is not easily explained by anything else. These costs are borne by the affected individuals, their families, their therapists, and the healthcare systems that treat them. They do not appear in Meta's quarterly filings.

The economic framework for thinking about this is externality pricing, and its application to attention extraction is both straightforward and radical.

Externality pricing works by asking: who pays for the consequences of this activity, and are those consequences included in the price of the activity? When they are not — when the consequences are paid by people other than the producer and consumer who generated them — the result is overconsumption of the damaging activity. The factory produces too much pollution because it does not pay for the river damage. The attention company captures too much attention because it does not pay for the cognitive damage.

If attention companies were required to internalize even a fraction of the costs they currently externalize, several things would happen. First, the advertising-based business model would become less profitable, because the effective price of engagement-maximized content would have to include a levy corresponding to its documented cognitive and social costs. Second, design incentives would shift: if you are paying for the mental health consequences of your design decisions, you have a powerful reason to make different design decisions. Third, the relative competitiveness of attention products designed for genuine user benefit rather than engagement maximization would increase, because the artificial cost advantage of extractive design would disappear.

The numbers are staggering when taken seriously. Global mental health costs attributable to social media run into hundreds of billions annually even on conservative estimates. Political dysfunction costs — the degradation of legislative capacity, the election of demonstrably worse leaders, the increased difficulty of passing consequential legislation because the public information environment has been radicalized — are harder to quantify but almost certainly larger. Productivity costs from reduced cognitive capacity are harder still to measure, but consider that the combined market capitalization of the companies most dependent on attention extraction is measured in trillions; the economic value of what they have degraded is at least comparable.

The objection that correlation is not causation is worth addressing directly, because it is the primary defense the industry deploys. The correlation between social media adoption and mental health deterioration is now established in multiple countries, multiple age cohorts, multiple longitudinal studies, using multiple methodologies. The causal mechanisms — social comparison, sleep disruption from screen exposure, reduction of in-person social activity, exposure to algorithmically amplified distressing content — are independently documented. The argument from uncertainty has become a sophisticated form of denial that resembles, in its structure, the tobacco industry's long campaign to maintain "scientific controversy" about smoking and cancer.

There is also a civilizational argument that goes beyond the economic. The kind of thinking that civilizations most need at this juncture — careful reasoning about existential risks, deliberate evaluation of complex policy tradeoffs, patient deliberation across genuine disagreement — is precisely the kind of thinking that the attention economy systematically degrades. We are facing problems that require more of humanity's best cognitive capacity precisely as the dominant information infrastructure is optimized to reduce it. The opportunity cost is not merely economic. It is civilizational in the strictest sense.

What would bankruptcy of the attention economy actually look like? Not necessarily the literal collapse of large companies, though that might follow in some cases. More likely it would look like a forced restructuring of the business model analogous to what happened to tobacco: continued existence, but with dramatically reduced profitability, constrained by regulation and liability, forced to operate in ways that do not impose unconstrained costs on the public. The advertising-maximization model, which generates revenue by maximizing engagement regardless of the effect on the user, becomes legally and financially untenable. What survives is a different set of products: subscription models that align company revenue with genuine user benefit, platforms designed for genuine connection rather than engagement extraction, information services that optimize for comprehension rather than time-on-site.

This is not utopian. It is the predictable outcome of applying standard economic logic to a market failure that has been running uncorrected for two decades. The political difficulty is not analytical; it is that the companies generating the rents have used those rents to purchase influence over the regulatory processes that would constrain them. This is also a standard pattern. It is resolved, eventually, by informed public demand that overwhelms the purchased influence — which requires, recursively, that publics be capable of sustained attention to the issue.

The attention economy's deepest trick is that it degrades the very cognitive capacity required to recognize and resist it.