In October 2024, a video of a poll worker in Bucks County, Pennsylvania destroying mail-in ballots began circulating on social media. Within hours, several hyperpartisan accounts had picked up and amplified the clip, framing it as proof of large-scale fraud. Local officials and federal cybersecurity agencies quickly determined the footage was fabricated, but it had already been cemented as truth. The commentary that followed blamed the “misinformed public” for blowing it out of proportion—a narrative that still exists as misinformation continues to infiltrate the informational environment.

But this is the convenient approach. Misinformation doesn’t spread because people are incapable of thinking critically, but because elites and platforms with reach and clear incentives legitimize it by capitalizing on an information ecosystem that amplifies outrage over accuracy. 

In a polarized polity like the U.S. today, identity determines which sources feel credible. And in an economy driven by attention, sensational content produces the engagement that platforms and political elites later monetize. When those emotion-based dynamics allow elites and platforms to manipulate public perception at scale, democracy falters.

To preservean ecosystem in which meaningful democratic assessment is even possible, we must recognize misinformation as the supply-side output of these institutional incentives. This helps explain why our existing interventions fall short: they target individuals without addressing the elites profiting from deception. As such, we must shift the focus upstream so we can build interventions targeting what makes deception so profitable in the first place.  

Why Misinformation Sticks 

Identity-driven reasoning and the asymmetric virality of sensational content help explain why falsehoods spread so efficiently: 

First, partisan identity functions as a powerful mechanism dictating who we trust. Researchers Shanto Iyengar and Sean J. Westwood revealed that affective polarization—intense hostility toward the opposing party—has grown so strong that partisans routinely prefer ingroup members even in nonpolitical settings. This aligns with social identity theory—the notion that people derive their self-worth from groups they identify with, leading to ingroup and outgroup formations. These findings help explain how misinformation travels as people instinctively trust identity-aligned content while rejecting opposing information. This means motivated reasoning isn’t always a matter of ignorance, as we automatically reason toward identity-protecting conclusions, making identity an easily-exloitable cognitive shortcut to perpetuate false, partisan narratives. 

But identity alone doesn’t explain the speed or extent to which misinformation spreads. The architecture of modern online platforms does the rest. MIT researchers find that false news spreads “farther, faster, deeper, and more broadly than the truth” on social media platforms because novel, emotional-laden stories trigger more engagement. This is a structural feature of platforms that relies on human attention to privilege content, eliciting quick, almost visceral reactions. Eventually, sharing misinformation becomes a mindless product of inattention in an interface that rewards factors speed and social gratification over reflection. This design produces a system in which even well-informed and -intentioned users can amplify misinformation. 

Importantly, this misinformation-polarization feedback loop does not place millions of citizens as the root. Studies show that the people most responsible for sharing large quantities of political misinformation constitute a subset of “superspreaders” like political influencers. Essentially, the public engages with misinformation not because it chooses to, but because identity cues and top-down incentives eventually culminate in an environment optimized for emotional partisan affirmation, allowing elite-generated misinformation to flourish with minimal resistance. 

The Elites

We’ve established that misinformation persists because elite institutions and actors have powerful incentives to manipulate public opinion. And once partisan identity and platform design create a public primed for emotionally congruent and novel content, the information environment becomes a marketplace, and misinformation one of its most valuable commodities. 

Political elites, especially, understand that influencing perceptions of reality is perhaps just as powerful as policymaking, with the former being far more efficient and pervasive. Elite actors like political candidates, elected officials, party strategists, and high-profile political surrogates, possess both the reach and credibility needed to inject falsehoods into ingroup networks with immediate effect. Research adds that when political actors validate misleading narratives—even if it’s just by mentioning it—their supporters rapidly follow suit regardless of prior beliefs, making elite cues an influential heuristic. Considering this with additional findings that misinformation consumption tends to cluster heavily among individuals already ingrained in partisan media ecosystems, it’s not entirely unreasonable to assume they were likely influenced by elite messaging. Validating misinformation that perpetuates partisan online discourse not only establishes the content’s credibility itself, but also grants political elites the most significant element to uphold their image and gain campaign support: attention. In 2021, The Washington Post recorded 30,573 false or misleading claims by then-president Donald Trump—most of which took place on social media accounts, significantly influencing national public opinion. As elites continue to legitimize misinformation for personal gain, they lose their credibility as reliable sources of information.

But if elite political actors endorse the lie, platforms determine whether it succeeds. Social media platforms rely on engagement for profit, and misinformation is exceptionally good at generating engagement because it reliably triggers anger, fear, and surprise. Surveillance capitalism, a business model where companies collect users' personal data to create "prediction products” for advertisers, explains how online platforms profit from capturing and predicting user attention, giving emotionally charged and identity-threatening content a structural advantage in the informational marketplace. With about half of U.S. adults reporting visiting platforms like Facebook, TikTok, Instagram, and YouTube daily, it’s safe to say this tactic works.

These advertising-driven revenue models also reward content that keeps users scrolling and sharing regardless of truth value. In fact, the 2022 annual advertising revenue across six social media platforms from users under age 17 alone amounted to $11 billion. For social media platforms, misinformation is only profitable: sensationalism draws attention (better if a political figure validates it), attention yields clicks, and clicks bring revenue. The fact that misinformation proliferates through inattentive sharing is even more troubling after recognizing that platforms have financial motivations to cultivate inattentiveness. 

Political figures benefit financially, too. Outrage-based fundraising is one of the most powerful tools in modern campaigning. Strategically drawing attention to misinformation and creating false narratives about partisan electoral threats or crises generates fear—and fear is extraordinarily effective at opening people’s wallets and expanding donor lists. 

The result is a closed-loop system: elites produce misinformation for personal advantage; platforms amplify it to profit off of user engagement; and partisan audiences reinforce it to signal party loyalty. To top it off, even subtle algorithmic tweaks can dramatically shift what information users encounter. Researchers reported a nearly 50% increase in hate speech following Elon Musk’s acquisition of Twitter in 2022. Since then, the platform has shifted dramatically, including the rise of nationalistic accounts like Inevitable West, which calls itself a “Defender of Western values and culture,” and algorithmic changes propagating right-wing posts—proving how platforms can be engineered toward a particular narrative. These powerhouses have built an algorithmic architecture that monetizes outrage and amplifies sensational headlines, leaving citizens to navigate an environment designed to mislead them. 

Why Traditional Interventions Fail

When misinformation is this profitable and this structurally ingrained, asking citizens to fact check content and resist emotional reflexes is nowhere near sufficient. The supply-side incentives and rewards are strong enough to overwhelm any individual-level remedy—it’s why our existing interventions rarely last long.

Fact-checking, for instance, relies on existing misinformation. But by the time there’s enough accurate information for a correction, the lie has already saturated networks, and any corrective effects struggle to gain traction in an environment that favors false narratives. The impact of fact-checking also “tends to be limited to the facts that are being corrected,” often “[turning] participants into passive recipients of specific corrections and thus failing to markedly enhance their skills,” further proving its inability to rework human cognition.

Similarly, media literacy campaigns assume that individuals share misinformation because they can’t distinguish fact from fiction. But we know that most users who share misinformation aren’t making deep epistemic judgments—they are sharing for social rewards, or doing so mindlessly as identity cues paint certain posts to be trustworthy regardless of the content itself. Teaching people to inspect URLs or evaluate sources, while important, doesn’t meaningfully alter what incentivizes partisan signaling in the first place, and neither does it change the platform’s architecture. Media literacy also simply cannot neutralize the engagement-based structure of social platforms, as misinformation travels six times faster than the truth, thus leaving even highly informed users victims to a system designed to elevate emotionally salient content.

Prebunking and accuracy prompts, show promise in laboratory conditions, but will struggle to keep up outside. These techniques work by forcing individuals to slow down and consider what makes a post accurate before sharing. But their effectiveness depends on attention and motivation—exactly what misinformation is designed to override. Accuracy prompts are proven to reduce the sharing of misinformation in controlled settings, momentarily shifting users’ attention to checking for accuracy. But this shift rarely lasts once people return to their emotionally arresting algorithmic feeds. Relatedly, while such accuracy prompts can alter beliefs in the moment, they fail to alter the emotional aspect that made the misinformation appealing in the first place, further surrendering long-term promise. And although prebunking improves accuracy judgments, it risks teaching people how to spread misinformation, which cannot compete with the volume or strategic repetition of elites that keep adding fuel to fire. 

This doesn’t mean citizen-level interventions are useless. It means they are structurally insufficient at countering the drivers that generate misinformation at a deeper systemic level. Fact-checking aims to correct downstream effects, media literacy tries to strengthen downstream processing, and accuracy nudges and prebunking efforts aim to reinforce downstream cognition. The upstream forces (e.g., political incentives, monetization structures, algorithmic platform designs) that determine what enters the information stream in the first place go ignored, creating an environment where even the most well-designed citizen-focused interventions will struggle. To address this crisis, we must therefore evaluate interventions not by their ability to educate people, but by their ability to effectively disrupt institutional incentives that reward deception. 

The Real Cost

Seeing the misinformation crisis as a structural feature reveals its impact and urgency as it constantly reshapes the conditions under which sound democratic judgment is even possible. 

Democracies depend on healthy disagreement about values, not basic reality. When elites strategically promote misinformation, they eliminate steady deliberation as citizens converse without a shared factual baseline, distorting citizen-level discourse and accountability. Elections function best when voters can evaluate leaders on the basis of outcomes. But when political actors can obscure failures or invent threats, people judge them not on their accomplishments, but on the narratives they manufacture. Those already embedded in partisan ecosystems interpret events through elite-curated lenses, further enabling politicians to evade responsibility. 

Broadly speaking, perpetuating false political narratives for profit undermines the very institutions integral to democracy. One fabricated video, like Bucks County’s, can be the catalyst in casting suspicion on local officials and basic legal procedures. When elites validate such information in any capacity, they inadvertently amplify narratives that force our institutions to have to fight to preserve their legitimacy. Misinformation is no longer a mere threat to factual accuracy, but a constraint on essential democratic elements and processes.

What Now? 

If misinformation endures because the system rewards it, then the most impactful interventions have the burden to target those rewards. A starting point is raising the cost of elite deception. False narratives rarely get traction without elite permission, and when high-visibility political actors fail to repeat them, their reach collapses. Establishing norms that penalize deception, be it through journalistic scrutiny or reputational cost, will do more to disrupt misinformation’s spread than any individual-level educational campaign. 

Another avenue involves targeted platform redesign. If virality fuels misinformation, then slowing its distribution can truly alter what reaches the public. Adding friction like de-emphasizing engagement in algorithmic ranking systems, labeling manipulated media (which Instagram employed in 2019 with some success before its parent company, Meta, discontinued the feature earlier this year), and establishing consistent transparency requirements can help shrink the informational asymmetry that currently advantages deceptive actors. Even modest changes to ranking systems can meaningfully shift what information reaches the public. 

None of these are cure-alls, and currently, most are challenging to even fully define, let alone achieve. But they all share a crucial logic of intervening at the level of the elites and their incentives—not the public. While it may seem challenging, we can and must make systemic changes if we wish to preserve democracy. We cannot keep treating the misinformation crisis as a byproduct of failures in individual-level judgement when ordinary citizens are being forced to navigate strategic conditions built to profit elites and platforms by manipulating their judgments. 

The real danger doesn’t lie in a “misinformed citizenry” but in the political environment we have created and enabled where the truth no longer holds any comparative value. If democracies require an informed public, they also require institutions that make the truth worth perpetuating. Until we flip the existing narrative and directly tackle the incentives that make misinformation profitable, citizen-focused solutions will keep losing to a manipulative system that makes misinformation both efficient and rewarding for the elites—turning democracy into collateral damage while doing so.