For Expertise to Matter, Nonpartisan Institutions Need New Communications Strategies

For most of the twentieth century, institutions whose authority rested on nonpartisan expertise influenced America’s public sphere through a top-down communications model. Think tanks, modern universities, and social and advocacy organizations produced writing and research aimed at policymakers and elites. Public communication was seen as the last stage of the process. Press releases alerted a relatively manageable set of high-profile outlets to new writings. Carefully crafted content passed through media gatekeepers who determined what deserved public attention. Through op-eds, Sunday talk shows, and briefings, experts could bring their work to the attention of politicians, executives, and other decisionmakers who could act on their ideas. The public—especially the educated and politically engaged—could be reached through a relatively small number of channels. 

Even as cable news expanded and partisan outlets like Fox News emerged, this approach held: Institutions created, gatekeepers validated, audiences consumed. The top-down, centralized approach rewarded polish, institutional prestige, and insider relationships with gatekeepers. A high degree of shared trust gave credence to the choices of the gatekeepers; legitimacy was communicated through successful participation in this hierarchy.

That system has collapsed. 

Today, public attention flows through a far more diffuse, competitive ecosystem—one where influence is shaped by networks rather than hierarchies. The old gatekeepers have been replaced by new ones: algorithms that curate content, high-follower social media accounts that influence what goes viral, and deeply engaged niche creators who enjoy immense legitimacy within their communities. The newly influential are not simply broadcasters at the top of a different hierarchical order—they determine what content matters in conjunction with active, participating audiences. Legitimacy is now conferred on those who master resonance, immediacy, emotional connection, and authenticity. 

Organizations and scholars grounded in fact-based argumentation—and the philanthropists and advocates who support research-backed policy influence—must grapple with this shift. There are, of course, still influential people who consume traditional media. Reports and op-eds still matter in elite technocratic circles (especially if they lean left). Headlines from legacy media show up in TikTok videos and social media newsfeeds. President Donald Trump is still significantly influenced by Fox News, even though he also engages with his social media audience. Yet even politicians and policymakers who favor major newspapers or television channels for their own information are influenced by, and must reason with, constituencies who inhabit a fundamentally different media ecosystem.

Institutions working on issues that depend on mass public support must adapt to today’s media environment—or risk irrelevance.

Public attention has shifted decisively. Institutions working on issues that depend on mass public support must adapt to today’s media environment—or risk irrelevance. Too often, scholars and advocates have tried to engage in new media using old media habits that are ill-suited to the medium. Some believe themselves exempt from having to engage on new media at all, assuming that esoteric or technical fields insulate them from this transformation in influence. But recent history—on issues ranging from public health to climate to foreign aid to security policy—shows how quickly expert domains can become contested online battlegrounds. Institutions unprepared or slow to engage have repeatedly been outflanked by bad-faith actors who were weak on facts but had superior communication tactics.

To avoid being sidelined just when they are needed most, experts and nonpartisan analysts must rethink not just their channels of communication but also their theory of influence.

In this paper, we discuss several aspects of how influence has transformed and how institutional and expert communicators might adapt. The first section explains the transformation of the media ecosystem and four major shifts that led legacy media to lose ground—including significant alterations to how credibility is conveyed. The following section discusses how social media influencers with large audiences are not simply new gatekeepers broadcasting to their followers but are more akin to community conveners, working with their audiences to make sense of the world and decide what is true together. The third section discusses how institutions and experts are largely absent from this process and why that matters for democracy and consensus. The concluding section offers five strategic recommendations for institutions and philanthropists to begin the urgent process of engaging with this new media environment.

How the Media Environment Has Changed 

Traditional forms of media such as newspapers, magazines, television, and cable news—collectively known as legacy media—still matter to policymakers, to world leaders, and to the audiences who enjoy them. But that segment is shrinking and is increasingly demographically narrow. It has already lost many young people, those who lean right, and a majority of Americans who are not seeking out news at all. Americans’ changing media habits result from four significant shifts in media production and public trust.

Personalities, Not Institutions

Legacy media is hierarchical, and it depends on large teams and infrastructure for production and dissemination: newsrooms of reporters, publishing houses full of editors, and academic presses supported by peer review networks. By contrast, claiming the mantle of media today merely requires an internet connection and an audience (though as we discuss later, offline networking and relationships with other influencers help build that online audience). This shift has moved influence away from organizations and toward individuals. 

Independent media-of-one figures may run Substack newsletters, podcasts, and blogs. Some are well-known journalists who have left institutional homes to strike out on their own; others are ordinary people who simply begin writing publicly on the internet. Many amass significant followings: Journalist Matthew Yglesias had around 18,000 paying subscribers in October 2024 and was earning over a million dollars a year after Substack fees and operating costs.¹ Letters From an American, a Substack newsletter written by history professor Heather Cox Richardson, has tens of thousands of paid subscribers and 2.6 million overall followers.² Larger partisan-niche outlets like The Bulwark operate with slightly bigger newsrooms of multiple writers and researchers, using low-cost distribution infrastructure—but even these are only lightly institutionalized, with audience engagement largely driven by the personalities of their star contributors.

This shift is not limited to entities that identify themselves as media. A growing number of content producers present simply as themselves. These individuals are typically called “creators” or “influencers”—terms that encompass a wide range of roles including streamers, podcasters, video essayists, meme accounts, journalistic explainers, and more. They span every topic and aesthetic: wellness coaches, climate activists, relationship commentators, historians, economists, comedians. A small but increasing number are experts who are experimenting with new mediums, such as professors translating courses into TikTok snippets. Influencers most commonly speak to their audiences not as “the media” but as fellow members of a community of interest or identity: just an ordinary mom, talking to you from her kitchen table about parenting struggles. But beneath the casual tone and relatable setting lies a follower count that, for prominent individual influencers, can be many times larger than the number of primetime CNN viewers.

Reach and Relevance

Unlike legacy media, most influencer-driven online media is free to the consumer, supported through ads, sponsorships, and revenue-sharing models. Traditional media, by contrast, often sits behind paywalls. Findings from Pew Research Center indicate that just 17 percent of Americans paid for news last year, and only 1 percent of Americans paid if they happened to hit a paywall.³ In other words, the New York Times, Washington Post, Wall Street Journal, National Review, Foreign Affairs, Atlantic, and nearly every other legacy media title that lies behind a paywall is consumed by a small, specialized subset of the population. Some people try to get the same information from various free sources, but with X and other social media platforms often limiting links⁴ and connections⁵ to news outlets, these secondary sources are themselves quite limited. Consequently, public intellectuals and experts who publish only through legacy media outlets are engaging a narrow, affluent, and educated demographic that is increasingly talking to itself.

Public intellectuals and experts who publish only through legacy media outlets are engaging a narrow, affluent, and educated demographic that is increasingly talking to itself.

The reach of the top performers in the new media ecosystem dwarfs the mass media channels that came before it. The information source with the largest audience as of this writing is not the New York Times, with 11.7 million subscribers,⁶ or the Wall Street Journal, which has approximately 4.3 million.⁷ It is not CBS, with its 5.7 million primetime viewers,⁸ or Fox News, with its 2.5 million.⁹ It is a man named Jimmy Donaldson, better known as MrBeast, who has over 400 million YouTube subscribers and nearly 120 million followers on X.¹⁰

Content creators and influencers like these are now shaping public conversations. They produce content across an extensive array of formats ranging from short-form TikTok videos and Instagram Reels to long-form YouTube videos, podcasts, or live streams on Twitch. They often maintain interactive community spaces on Discord or engage with their fans on Reddit. Many also have audiences on the growing proliferation of platforms dedicated to serving niche audiences who align around ideology, such as Rumble or Truth Social for the right and Threads or Bluesky for more left-leaning audiences. The fact that creators have captured attention across a wide range of audiences and mediums gives them a significant impact on culture and the dissemination of information. Shifts in how legitimacy is attained mean that their opinions on news are taken seriously by their followers, even though they do not present themselves as experts or journalists.

Entertainment as a Source of News and Ideology

At first glance, MrBeast’s numbers might seem irrelevant to policy analysts or scholars. He gives away large sums of money in attention-grabbing stunts, which does not seem to directly compete with news or informational programming. He and others who command massive attention in the online and social media worlds—Joe Rogan in podcasts, Hasan Piker in streaming, Emma Chamberlain on YouTube, Kylie Jenner on Instagram, Charli D’Amelio and Khaby Lame on TikTok, and alt-media conglomerates like Barstool Sports across platforms—are first and foremost entertainers, not news providers. 

But today, most Americans do not seek out news. In August 2022, a Pew Research Center survey found that only 38 percent of Americans claimed to check news regularly—a sustained decline from 51 percent in March 2016.¹¹ Instead, they now encounter issue-related or political information ambiently, often embedded within the entertainment and lifestyle content they consume, on platforms where they already spend their time.

Influencers are central to this ambient news model. The creators who have the most impact on shaping public understanding of policy, science, and social or political issues today are often not political commentators or subject-matter experts at all.¹² They are lifestyle influencers, comedians, streamers, parent podcasters, and wellness creators. Their audiences come for aesthetic appeal, practical advice, community connections, or entertainment. Followers build trusted and at times parasocial relationships with these creators, who present themselves as fellow members of a community or identity, not as professional media. Audiences are exposed to policy ideas and cultural narratives incidentally rather than intentionally. 

Following the 2024 U.S. election, many philanthropists looked at the gap in audience size and engagement between influencers and legacy media and processed it as a need to fund the creation of more political explainer content or to support civic media initiatives that spoke to political questions. Some partisan players wanted to create a “Joe Rogan of the left”—missing the fact that Rogan’s mass appeal is because most of his content is cultural, not political. It is true that political content creators can capture politically engaged audiences. But explicitly partisan, ideological, or policy-focused content rarely appeals to the much larger, politically disengaged majority, including the large segment of the public that is avoiding news. 

Content creators may reach audiences through humor, lifestyle, or community—but over time, they also transmit values, narratives, and political frames.

Content creators may reach audiences through humor, lifestyle, or community—but over time, they also transmit values, narratives, and political frames. For institutions that seek to shape public understanding, ignoring these figures as unserious is no longer viable. They are already shaping the public conversation.

Getting information into the attention space of most Americans—and having them trust that information—is a fundamentally different activity than it was for the previous century. It requires building connection and trust first and layering in policy content later.

Authenticity as the New Legitimacy

What sets influencers apart is their mastery of authenticity, immediacy, and emotional resonance.¹³ This is foundational to why their audiences trust them. This shift coincides with declining trust and a credibility gap in legacy media, driven partly by partisan polarization. Among traditional media outlets, Republicans trust only Fox News, while Democrats rely on a broader range of outlets.¹⁴ Nonpartisan experts seeking to reach the public may be unaware of the ideological narrowness of the readership of many legacy media sites. (The Atlantic’s readership, for example, leans more Democratic than any of the other thirty major sources the Pew Research Center studied, including MSNBC and the Guardian.¹⁵)

But the changes to which voices are viewed as credible goes beyond partisanship. In past years, the mark of approval from hallowed institutions like the New York Times or a Sunday talk show conveyed that the content had institutional approval, had likely gone through a fact-checking process, and was therefore legitimate and true. But as trust in institutions has declined, many Americans began to view polished institutional statements with suspicion—seeing them as potentially manipulative, overly cautious, or influenced by corporate or political interests. Audiences today frequently interpret immediacy, messiness, and raw emotional expression as signals of truthfulness.¹⁶ They gauge credibility based on perceived sincerity, conveyed by a casual conversational tone or spontaneous, relatable presentation. Thus, a creator speaking informally from their car or home feels more credible to many than a university president delivering precisely crafted remarks. 

This emotional reading of credibility has deep psychological roots. As persuasion expert Robert Cialdini and others have shown, people are more likely to be persuaded by those who seem relatable, warm, and familiar.¹⁷ Informal communication styles amplify this effect, making messages feel more trustworthy. But while the public often sees immediacy and casual presentation as signs of authenticity, that does not mean influencers are free from outside influence. Influencers who do not rely on audience subscriptions often earn their income through advertising, product placements, or sponsorships—sometimes without disclosing these arrangements. Their business models can be opaque and potentially susceptible to undue influence.¹⁸ Yet many viewers nonetheless see these creators as more trustworthy than institutional media, reflecting a populist skepticism of institutions that cuts across the political spectrum.

The most popular creators excel at emotional connection. Influencers blend accessibility, perceived credibility, and community representation into personas that confer distinctive persuasive potential.¹⁹ Because they compete for attention—which converts to views, followers, and potentially revenue—in a very crowded ecosystem, many initially focus their content on a niche: an identity-based community, subculture, or specific hobby or interest area. (Some are explicitly political or partisan, too, of course, catering to communities who are still seeking out political news or ideological camaraderie.)²⁰ Influencers often come to serve as anchors for their communities, speaking not only to audiences but for them. They are seen as embodying the values and identities of their followers, turning digital networks into spaces of shared purpose and mutual support.²¹ Online niche communities—with their own symbols, in-jokes, and styles of expression—can foster a strong sense of belonging, community, and collective identity.²² Some of these niche communities can deliver immense follower counts; others boast more modest numbers. Influencers with small follower counts but high community resonance are sometimes called “microinfluencers” or “niche internet microcelebrities”; the legitimacy and trust they enjoy within their communities is significant enough that they are influential in ways that should not be underestimated based on numbers alone.²³

Influencers win attention and legitimacy via authenticity, relatability, and direct audience connection. Institutions and experts, by contrast, generally continue to signal credibility via authority: facts presented alongside credentialling markers such as educational degrees, institutional brands, or peer review. Many have chosen not to participate in the new attention arena; some are frustrated or discouraged by the short audience attention span and fleeting nature of the new forms of media, so they write it off. There are limitations to conveying complexity in the character limit of a Bluesky post or the sixty-second length of an Instagram Reel. But podcasts do not suffer these limitations: They are often one to three hours long and can delve deeply into topics. However, that medium expects guests to chat on a variety of issues, straying from areas of deep knowledge or talking points crafted for a television-based world of sound bites. One of the most difficult challenges of moving fact-based analysis into the social media era is neither format nor length, per se, but that the requirements to succeed on new media platforms are almost precisely the opposite of what successful experts, politicians, and advocates have trained themselves to do: maintain an institutional tone of formal detachment and stay ruthlessly on-message at all times.

The requirements to succeed on new media platforms are almost precisely the opposite of what successful experts have trained themselves to do: maintain an institutional tone of formal detachment and stay ruthlessly on-message at all times.

Social platforms reward interactive, casual, visual, and emotionally engaging creators and content; traditional outputs are formal, static, and detached. Institutions and experts struggle to adapt precisely because the skills needed to thrive—relational communication, conversational fluency, responsiveness, and emotional resonance—are largely absent from traditional educational and institutional training. Being spontaneous and relatable is an innately different skill set compared to the polished persona and ability to deliver a sound bite that the broadcast era rewarded. 

Ultimately, influence in today’s media landscape hinges not on hierarchical validation from established institutions but on reach, resonance, relevance, and authenticity. Institutions aiming for meaningful public engagement must adapt—not merely in terms of content but also in their approach, style, and delivery.

Maximizing Buy-In and Reach Through the Algorithm, the Audience, and Other Influencers

Influencers are not simply broadcasters atop a new online hierarchy. The online and offline relationships they build with other influencers help distribute their content across loosely networked communities. They establish frames and memes together with their audience, jointly riffing in ways that generate community buy-in.²⁴ Influencers, in other words, do not stand alone as celebrity personalities—their content generates engagement and spreads because of relationships with their audiences and other influencers. Successful social media engagement is not simply about getting one’s message into a series of vertical videos or speaking on a long-form podcast. It requires ongoing engagement to build an audience and to network across different influencer communities so that one’s voice achieves organic expansion.

But today’s creators do not just serve their audiences—they also serve “the algorithm.” (While social platforms leverage a variety of algorithms, “the algorithm” is commonly shorthand for the powerful recommendation and ranking systems that determine what content surfaces in users’ feeds.) Just as experts of the past had to know the news cycle, today’s creators develop fluency in what the algorithm rewards.

The platforms are optimizing for retention, satisfaction, and time-on-site: They are curating content to keep users coming back, via a combination of hooks and emotional resonance. Algorithms such as the trending algorithm incentivize users to pay attention to particular topics; others suggest accounts for users to follow. Some platforms, such as TikTok, pay little attention to who users follow and instead primarily promote content according to what they think the individual user will enjoy, heavily weighting recent content.

Creators must therefore navigate two audiences simultaneously: the platform’s users, and its ranking systems. Their content must appeal to an algorithm’s logic while also resonating with their followers or their broader target audience. Early likes, comments, shares, and watch time help signal to the algorithm that a piece of content is worth wider promotion to more potential viewers. Attention is earned one click at a time. And so, aspiring creators study what kinds of thumbnails, titles, and formats platform algorithms promote.²⁵

Across social media, the incentives favor content that is emotionally resonant and immediately compelling. The ecosystem rewards creators who post frequently and react quickly—qualities that do not always mesh with deep research or reflective thinking. Instead of editorial rigor or factuality, the dominant logic is interaction. Nuance is often lost. 

Across social media, the incentives favor content that is emotionally resonant and immediately compelling. Instead of editorial rigor or factuality, the dominant logic is interaction.

But amplification does not happen through platform algorithms alone. It also emerges through human networks. In the broadcast media ecosystem, institutions built relationships with editors and other gatekeepers—in the new system, relationships with offline communities and with other influencers are just as important. Influencers form ecosystems of mutual promotion: Just as old television specials would sometimes bring famous characters from different series together to boost viewership, a podcast with a relatively well-known influencer might bring an even better-known personality on as a guest, helping both shows build viewers from the other’s audience.²⁶ Some of this is organic: communities engaging in mutual support to help a cause, for example. Sometimes it is strategic, driven by influencers, agencies that represent them, or partisan networks. (Occasionally it is also inauthentic, such as when bots or undisclosed paid amplification are involved.) 

Amplification networks consist of more than just other influencers. A critical distinction between the hierarchical legacy media model and today’s networked social media ecosystem is how much audience participation matters.²⁷ In the prior model, institutional actors—universities, think tanks, government agencies—talked to their audiences. In the social media ecosystem, creators talk with their audiences—often in real time through live streams, comment replies, “ask me anything” segments (AMAs), and TikTok stitches. They respond to feedback and evolve their narratives accordingly. Viewers become collaborators, shaping the flow and framing of content. Live streams and AMAs replace press conferences. Platforms like Twitch and Discord create small digital town squares where creators are both host and participant, reading chats, referencing community in-jokes, and reacting to live sentiments.

Because influencers engage closely with their followers, the audience influences the influencer. Audience feedback actively shapes what many influencers produce. The result is a cocreation process: Influencers and their followers collectively build and spread narratives. Community members often serve as sources of both information and inspiration—surfacing ideas, observations, or content that influencers then amplify. This often happens when an influencer pulls a post from a smaller account in their orbit and boosts it, such as quoting a follower’s post on X. That act of amplification can push a piece of content into broader visibility, prompting responses from other influencers and spreading it across platforms. What begins as a niche observation can quickly evolve into a cross-platform wave of commentary, taking an idea from a subculture into the mainstream.²⁸ 

This networked creation and amplification process is highly engaged and responsive. Narratives, rumors, and even propaganda campaigns can arise dynamically and collaboratively in real time and are shaped by continuous feedback from the audience and platform algorithms. Feedback from the online crowd lets influencers—and, in the realm of politics, aligned political elites—know how their audience feels about a topic immediately. These dynamics fundamentally alter who sets agendas, determines truths, and shapes public perceptions.²⁹

Audience members are not just consuming—they are contributing. Members of these communities believe information because they are cocreating it.

One of the clearest outcomes of this participatory model is buy-in. Audience members are not just consuming—they are contributing. They commit by boosting a trend, defending a take, or pushing a clip into new contexts. Always-on conversational formats turn feeds into communities with relational trust, loyalty, and shared identity.³⁰ Members of these communities believe information because they are cocreating it via shared acts of participation and through authentic relationships with the original creator and others in their community.³¹ This is why truth in the participatory media landscape often looks different than truth in a peer-reviewed journal. It is shaped less by argument than by affirmation. A message feels true because the community says it is—because trusted voices repeat it, in familiar language, within a shared emotional world. People do not just believe what they see. They believe what they help spread.

This dynamic has produced a new architecture of influence—a triad of the influencer, the algorithm, and the crowd that is outcompeting the hierarchy of the broadcast era.³² Creators pull ideas from their followers—memes, jokes, and evolving narratives, among others—and fold them into content. Algorithms both surface and shape what circulates. Crowds provide the amplification, emotional cues, and validation that signal what matters—and they influence the influencers. Together, this loop can transform an isolated post into a collective narrative. Legacy media, in turn, may report on what is framed as “some people online” are saying, carrying the ideas to those who do not participate in the ecosystem themselves.

This is how public opinion is shaped today—not through top-down directives but through a dynamic, participatory process rooted in collective behavior and networked amplification.

This networked distribution, remixing, and sensemaking process is why someone like Joe Rogan—even as a central node in the network—does not just succeed on his own merit. His influence is multiplied by appearances, reactions, and downstream content from a sprawling set of adjacent creators and followers. Efforts to build a “new Joe Rogan” miss the point: What matters is not only Rogan the individual but also the network that sustains and amplifies him. Influence today is not the product of a single voice approved by gatekeepers—it is the emergent property of a system.

AI and Social Networks

Artificial intelligence (AI) is rapidly transforming the information environment once again. In recent decades, research-backed institutions and experts learned to optimize for a search engine–driven landscape, where they could rise in the rankings if they strategically structured titles and keywords. Ensuring that one’s report was the first or second thing to surface in a Google search, and was available for free, could lead to outreach from reporters or policymakers, discoverability among regular Americans, and traffic to institutional webpages.

Now, large language models are rapidly transforming search engines into answer engines. In recent months, Bain & Company has found that about 60 percent of users’ information queries end with them reading an AI summary at the top of the search page—no further clicks through to any website.³³ Users are not landing on institutional websites and exploring adjacent content, leading to significant traffic (and potentially revenue) declines. This is likely to become even more consequential as Google and other search engines roll out formats where their primary return is not a list of search results but an AI-written custom article attuned to the search question.³⁴

Chatbots have also become a source of preprocessed information; users ask a question, and the chatbot offers what appears to be a complete answer, after which users stop searching and may not wonder what might be left out. With these two changes, AI has begun to threaten the business model of the internet in much the same way that the online forum Craigslist undermined the economic stability of newspapers built on ad revenue.³⁵ The human-created content of much of the internet—including commercial news—is paid for via ad clicks that rely on eyes on pages, so this is potentially a transformative shift in terms of which outlets will survive.

It may also be transformative in terms of what is created: Those who want to shape public opinion will now need to learn how to get their content into the results conveyed by AI models—each of which is its own black box, sometimes even to the developers themselves. Just as social media leads to evolutions in content (for example, the rise of vertical video in response to platform algorithms promoting it), writing will likely evolve in response to AI “readers” and curators.

Public engagement with AI is transforming social media in unexpected ways as well. Users on X increasingly question claims made on the platform by asking the platform’s built-in AI, Grok, “@grok, is this true?”—treating the xAI chatbot as a referee and fact checker. This can be helpful if the bot returns accurate information; some early research even suggests that conspiracy theory believers may be more willing to reconsider their views when corrected by AI than by a human.³⁶ However, Grok itself has been embroiled in controversies in which it returned incendiary and false content—notably, after xAI leadership intentionally recalibrated it via the system prompt to “not shy away from making claims which are politically incorrect, as long as they are well substantiated.”³⁷ AI answer engines and interfaces more broadly have been known to hallucinate, returning inaccurate or misleading information in response to requests for simple factual information (for example, Google’s Gemini telling users to keep cheese from sliding off of pizza using glue³⁸).

The social dynamics are complex, and it is not yet clear how trust will transform in the age of AI. People tend to believe information that comes from trusted sources and that aligns with their worldview. If an answer returned by an AI model contradicts the influencers or communities with whom someone identifies, decades of psychological research suggests that this tendency toward motivated cognition is likely to lead those seeking information to reject the answer—and double down on the trusted human voice instead. In some cases, users are already building bespoke chatbots fine-tuned to reflect their worldviews and sharing them with others in their networks.³⁹ In an increasingly complex world where discerning whether content is even real feels challenging, AI may thus deepen trust in known social media influencers and other sources that offer reassurance and credibility.

Institutions Are Absent, and That Absence Matters

Communications strategies today are not just about getting the message right. They are about understanding the structure of information flows. In this ecosystem, influence belongs not to those who speak most precisely or factually but to those who understand where—and how—audiences are listening.

Institutions are frequently absent from this space. Those that are attempting to enter social media often do so as broadcasters, not as community conveners. Some institutions have begun institutional podcasts or are putting out information via vertical videos—but they frequently drop content onto a platform rather than opening a conversation with their audience or understanding what a platform’s algorithm will reward. This broadcast style is not optimized for virality and often does not spread beyond the existing reach of the organization. Meanwhile, traditional institutional trust signals—prestigious affiliations, credentials, and citations—often do not matter as much in these environments. The tone of listing off bona fides or presenting information formally is not optimized for platform recommender dynamics. Thus, when institutions do attempt to participate in new media—often via official accounts or press statements—the mismatch in tone, timing, and texture is glaring. The posts may be present on the platform, but they do not fit the style or resonate with the audience and are unlikely to spread.

Institutions frequently drop content onto a platform rather than opening a conversation with their audience or understanding what a platform’s algorithm will reward.

This is not just a tactical failure. It represents a structural misalignment between how institutions produce information and how today’s media environments shape opinion. Institutional content is often optimized for clarity, precision, or objectivity—values that mattered most to the broadcast gatekeepers and that are still sorely needed today. But in participatory spaces, trustworthy, shareable content is determined based on authenticity, immediacy, and relational credibility.

Creators are adapting dynamically to the desires of their audiences, building information environments that are stimulating, engaging, and often fulfilling—but not necessarily grounded in fact. Indeed, the dynamic call-and-response process of social media crowds and the speed of social media virality mean that even brazenly false narratives can calcify rapidly. Rather than being filtered out, fringe ideas may gain traction within specific niches or on specific ideologically segregated platforms via emotional salience and repeated exposure. A poorly sourced claim, once embraced by a community or boosted by elite influencers, can become resistant to correction. Attempts to rebut it from the outside can feel like an attack, triggering defensiveness rather than reconsideration. The issue is not simply one of facts or misinformation; it is belief entrenchment within insular niche groups and significant division across them. 

Institutions are at a distinct disadvantage in today’s fragmented media environment, where traditional press releases, formal reports, and sign-on letters from luminaries no longer reach most Americans. Public trust, once the product of decades of experts’ investment in institutions and education, can no longer be assumed. It must be earned through means of communication that are nearly the opposite of the methods learned by ascending the institutional ladder.

Social media fundamentally altered the shared facts, reasoned debates, and voices deemed worthy of influence that once underpinned twentieth-century liberal democracies. The expansion of perspectives is an undeniable good. Yet today, a glance through one’s Google News or Instagram feeds offers minimal insight into what others might be reading. In a personalized system, asking whether one’s news matches the majority’s means little when each person’s information preferences can be used to create bespoke realities. In the absence of a shared public sphere and amid rising distrust in institutions, populist narratives—often framed in terms of conspiracy, heroes, and villains—rally people to defend their in-groups against perceived threats from the other side. In such a landscape, influence accrues to those who show up and engage; when institutions do not, the frames and narratives are set without them.

Yet interventions that try to return to twentieth-century models of shared facts and institutionalized sources of legitimacy face an uphill battle. Philanthropy invested a lot in the late 2010s trying to get platforms to enact fact-checking and address poor quality information. But appeals to facts falter when shared reality and widely trusted sources are rapidly declining. Expertise fails to command respect if the universities experts attend and the institutions that validate their trustworthiness are themselves viewed as tainted or illegitimate. Trust in institutions has not diminished solely due to shifts in communication patterns; there have been very real failures by experts and political leaders, from the 2007–2008 financial crisis, to long wars that were not won, to aspects of COVID-19 pandemic policy. For experts to regain legitimacy and trust, they will need to account for failures and oversights. These failures, however, also opened the door to sustained bad-faith attacks. What is required today is a combination of accountability and engagement: showing up in the arenas where the public forms opinions, conveying accurate information, and contesting bad-faith narratives in real time.

The country needs fact-based institutions to be in the game, to conduct long-term research using shared methodological baselines, to separate signal from noise, and to operate with incentives beyond short-term popularity. Institutions help to make debate meaningful and action possible. It is imperative that truthful information and serious researchers master this new media environment so that they are participating in the conversation in the twenty-first century. 

The era of “we are the media now”⁴⁰ represents a profound diffusion of communicative power and decentralization of authority that was previously concentrated among institutional voices. Yet institutional and fact-based expertise is still deeply needed. For the good of democracy and society, institutions must evolve how they engage the public. This requires reimagined strategies built around genuine interaction, responsive narrative formation, and sustained community trust.

Strategic Recommendations 

Given how sensemaking now occurs in online communities—and the need to remedy institutional absence—institutions, experts, and the philanthropists that support them must assess what they are trying to do with their communications strategies.

If the goal is to influence either a technical issue that has not yet garnered public attention or a community that is highly educated, well-targeted reports, op-eds, short articles, speeches, and private convenings provided to key stakeholders still work. Philanthropy, business, and some fairly technical political activity will continue to be influenced in this traditional manner. (That is, after all, what we are doing with this paper, which is targeted at just such an educated and engaged audience.)

Supporting new news sites or ideologically driven creators is a useful strategy if advocates or philanthropists are trying to deepen the ideological commitment of a politically engaged audience. News sites can energize the small segment of the population that is seeking out news and is already drawn to ideological content. But they will not reach the majority of the public that is largely avoiding news altogether (or at best skimming it⁴¹). Subscriber-based sites will also lose Americans who will not pay. Thus, creating or amplifying ideological news is primarily a strategy of mobilization, not persuasion.

For organizations that are attempting to educate, persuade, or influence a public debate, a new communications strategy is needed. Some institutions have already begun experimenting with ways to reach this audience. For those looking for actionable, tactical ways to make this switch, here are five ways to begin.

Prioritize Building Capacity in Advance

Organizations, experts, and the philanthropists that support them must build the capacity to engage in the current media ecosystem long before their issue becomes relevant. Consider that it took just a few weeks of misleading public commentary on social media to justify destroying the massive infrastructure and investment that was the U.S. Agency for International Development—a process that was driven in large part by rumors, spread by influencers and online crowds, that generated outrage that was in turn used to justify the dismantling action the administration intended to pursue anyway.⁴² There was no networked counter-speech capacity available to push back against misleading claims. Two weeks is far too short a timeline to create a networked ecosystem to defend the role of international aid—or any other field of action that experts and philanthropists may deem important. Major pieces of policy are now moving with equal rapidity: A massive increase in the deficit, huge changes to healthcare policy, and cuts that threaten to end public broadcasting were enacted with such speed that there was nearly no public discussion. The reputational destruction of a bastion of philanthropic funding or a respected institution can happen just as fast. Indeed, the attacks on specific research centers, universities, and philanthropic entities have already begun.⁴³

It is essential to start building the capacity to speak to the mass public now. This cannot be a secondary strategy or afterthought; it is a critical component of nearly all philanthropic efforts, and most organizations in the fact-based space must engage now.

Create a Presence at the Individual and Institutional Level

Experts who are willing to participate directly in the public conversation should invest in building their own authentic social media presence, including through video content, live streams, and other interactive formats. Institutions and philanthropists should support this experimentation, offering trainings for those who are interested in learning and providing extra support to experts who turn out to be effective communicators in this new environment.

Meanwhile, larger institutions should consider building “creator labs” into their existing communications infrastructure. These would be small, in‑house teams that translate dense reports into shorts, live streams, and explainers. Such creator labs would relieve experts of some of the significant time burden of this work, even as they remained the personalities behind the content.

Leadership visibility matters, too. Senior officials should be trained and scheduled for regular “ask me anything” segments, “get ready with me” videos, or behind-the-scenes content that allows the public to see the human beings behind the institution. Presence builds trust; repetition builds familiarity.

Finally, institutions should experiment with online-offline hybrids that tie existing community gatherings—at schools, book clubs, or faith venues—to live streamed question-and-answer sessions with trusted local voices. This pairing anchors digital content in real-world relationships and allows information to circulate through both personal connection and social media reach.

Researchers who become more public will, unfortunately, be more likely to face some of the offensive commentary that has become part of the public discourse. Much of this remains in the realm of harsh or inappropriate feedback that can be filtered out with software or otherwise ignored. But institutions that encourage their experts to engage on social media should also support them with measures to prevent doxxing and other forms of harassment. Ounce-of-prevention actions should become standard practice, such as support for experts to enroll in digital monitoring software that removes references to one’s address and family prior to developing a social media presence. Philanthropists should consider providing a portion of grants for this use and investing in the broader institutional architecture that is growing to offer support to individuals and institutions that require more help.

Cultivate Partnerships and Collaborative Storytelling

The new communication landscape often does not privilege the skills that experts have spent a lifetime building. Recognizing this reality means institutions should embrace collaborative storytelling between those who hold deep knowledge and those with the cultural fluency and media reach to deliver it effectively. Experts may bristle at creators with less educational investment gaining prominence for spreading research and ideas that they did not originate—but partnering may be the fastest and most effective way to get information into the public conversation.

There are a handful of approaches to consider; we discuss three common ones here. Philanthropy should support testing a variety of collaboration types rigorously to understand which add value, while recognizing that the field is swiftly changing and more options may soon be available.

Institutions should embrace collective storytelling between those who hold deep knowledge and those with the cultural fluency and media reach to deliver it effectively.

One model that has become common is paying creators to place content, including content cocreated with outside parties. Various firms specialize in helping interested parties connect with creators who speak to specific demographics. Many of these collaborations are already happening, as advocacy organizations hire firms that pay creators to deliver messages in accessible, emotionally resonant ways. Such pay-to-play efforts are not unlike political campaigns or advocacy groups buying TV ads; the path from money spent to impressions earned is clear. While paid partnerships are a way to experiment with new media, they can be expensive, and growth is inorganic—the content stops when the money stops flowing. This model is unlikely to be a long-term institutional solution to the new media challenge.

A second model involves building real, sustained offline relationships between experts or institutions and existing influencers whose goals and audiences closely align. Funding this model requires investments in relationship-building infrastructure to help experts and content creators meet and find points of connection. These relationships are often two-way: Just as experts are looking to enter the new media ecosystem, many influencers value expertise and relationships with knowledge generators. Many influencers are already engaging large or relevant communities, but a lack of funding derails their natural interest in producing more content aligned with particular expert issue areas; additional support can benefit them as well.

A third model involves philanthropists and institutions helping to create, surface, and amplify new voices. These individuals may be young, values-aligned people who want to become influencers as a career, or new-to-the-field creators who are already operating in this ecosystem. They show real promise in social media and care deeply about the issues at stake, but may lack skills in their media formats, need help building connections and audiences, or would benefit from institutional ties. These are often the people who, with a small investment of support and information access, can grow into powerful narrative allies. They have the time, talent, and trust to build durable communities—but they need to be discovered and supported.

Both the second and third models require thinking about and supporting the career paths of influencers, who make their living from their work. There are many tactics that deserve experimentation. Some organizations are seeking out talented college students to cultivate into influencers in their fields, just as institutions offer internships with skill building and language training to cultivate the next generation of climate advocates, nuclear experts, or China hands. Fellowships could spur new voices whose paths afterward would be entrepreneurial, just like the majority of content creators are today. Alternatively, these new creators might be funded directly by institutions as part of a communications team. People who excel as communicators and audience conveners could be paid to be part of an institution and serve as the expert known to the public; Bill Nye the Science Guy does not do all his research himself but has served as a brilliant explainer and entertainer to the general public. This path might be helpful particularly as fields are experimenting with how to move fact-based knowledge into an entertainment-, identity-, and community-based ecosystem. Such communicators could alternately find a career path as part of values-aligned media groups that could be supported by philanthropists and institutions. 

In all cases, these partnerships must be long term. Algorithms reward consistency and familiarity. A one-off post or campaign rarely moves the needle. What works is sustained presence, ongoing collaboration, and storytelling that evolves with the community—not just with the news cycle.

Incentivize and Support Cooperation to Scale

The communications need is so vast and new that it cannot wait for each individual nonprofit to figure out how to help its communications department get it right. Philanthropy should invest in shared infrastructure across nonprofits in the same field and should incentivize grantees to work together and amplify one another. 

For instance, creator labs need not sit only at large institutions. They could also be a funded institutional ability within an operating foundation or a service offered to a field of grantees—just as venture capital firms offer services to portfolio companies.

Many of the best creators are passionate about a field or issue. An effective group of creators could be supported to operate with many environmental organizations, for instance, or to help women’s causes, or democracy. Collaborations between creators within an issue area are natural on social media—and unnatural to how nonprofit fields tend to function. Currently, nonprofits in the same field are often fiercely competitive, jealously guarding who gets credit for what work so that they can make credible claims to funders. We understand this reality, as well as the fact that some nonprofits believe others in the same field are wrong in their ends or methods—these are not simple problems, or mere issues of ego. However, this individualist operating model is at odds with what is needed to succeed in calling attention to ideas on social media, where fields—with funder support—must instead create mutual amplification networks. Funders have the greatest power to change this dynamic. They must incentivize such amplification and make it a key component of the work itself.

Nonprofit organizations should be encouraged and funded to work together to learn how to engage with the new media environment and share best practices. Nonprofits in the same field must become accustomed to supporting each other and amplifying one another’s work to increase the networked audience effect. Philanthropists should pay less attention to which singular organization seems to be engaging in social media—and much more to whether an organization is lifting up the field as a whole.

Measure for Trust and Engagement

Every institution and creator needs to determine whether they are hitting the audiences they want to reach and ensure that messages are truly resonating. Community participation metrics such as comments and shares are important, as are variables that measure trust. Philanthropists, meanwhile, should evaluate progress not only at the institutional level but also across the broader field.

To accelerate learning, foundations should prioritize transparent reporting of impact metrics and foster an adaptive culture that recognizes how rapidly the communications environment is evolving. One way to enable such learning is by supporting quarterly convenings of communication leads to share the results of their experiments: Creators, researchers, and funders could swap metrics, case studies, and failures, with summaries shared after each meeting. Participants could be encouraged to form dynamic, real-time groups to share information and swap ideas between meetings.

The goal should be to support continuous improvement and to encourage experimentation. It is critical to acknowledge and fix what is not working—not to assign blame, but to turn setbacks into opportunities to learn at the individual, institutional, and field levels. Reward bold attempts even when they fail. Dig more deeply into the reasoning behind traditional expected metrics like the number of high-level, paywalled publications: Is that audience right for the institution’s primary objective?

Conclusion and Call to Action

Institutions that want to shape public discourse can no longer rely on prestige, credentials, or gatekeepers to privilege their content or opinions. Influence today increasingly flows through networks, not hierarchies—through creators, algorithms, and participatory audiences who decide what is seen, shared, and believed. This shift has not just displaced old media models; it has upended the assumptions about how trust and legitimacy are earned.

Today’s information environment rewards a different set of skills: emotional resonance, responsiveness, community fluency, and a willingness to meet people where they are. The institutions best positioned to make an impact are those that recognize this shift—not just in tactics but also in values. 

If facts and expertise are to continue to be influential in this new public sphere, they must be communicated in ways that build connection and credibility. Experts must prioritize building trust through regular engagement with the public. That means investing in new messengers, new formats, and new relationships. It means treating communication not as something to do once a report is complete, but as a core, ongoing component of the work itself.

By embracing this new challenge, fact-based experts, advocates, institutions, and philanthropy can amplify impact, help rebuild trust in institutions, and drive more inclusive, participatory public discourse. In a networked world, institutions must not only speak. They must participate.

Acknowledgments

The Carnegie Endowment for International Peace would like to thank Alta Futures, the Ford Foundation, the William and Flora Hewlett Foundation, the Joyce Foundation, and the David & Lucile Packard Foundation for generous support that helped make the writing of this paper possible. The views expressed herein are the responsibility of the authors alone.

About the Authors

Renée DiResta is an associate research professor at the Georgetown University McCourt School of Public Policy, where she focuses on influence and propaganda, user-centric infrastructure for trustworthy digital publics, and institutional resilience. She was formerly the research manager at the Stanford Internet Observatory, housed at Stanford’s Freeman Spogli Institute for International Studies.

Rachel Kleinfeld is a senior fellow in Carnegie’s Democracy, Conflict, and Governance Program, where she focuses on democracies experiencing backsliding, polarization, violence, and other governance problems.