“The conscious and intelligent manipulation of the organized habits and opinions of the masses is an important element in democratic society.” — Edward Bernays, Propaganda
A recent Jewish People Policy Institute study used artificial intelligence tools to examine public statements by United Nations Secretary-General António Guterres. The study reportedly found a consistent pattern in the frequency and tone of his references to Israel. The finding matters, but its larger importance lies in what it reveals about modern politics: institutional language can now be measured.
For years, claims of bias inside international organizations were treated as matters of impression. One reader saw imbalance; another saw emphasis. One government saw hostility; another saw accountability. The argument often dissolved into competing sensitivities. Artificial intelligence changes that debate completely: Frequency, tone, repetition, thematic emphasis, and omission can now be studied across large bodies of text. Patterns once dismissed as anecdotal can be analyzed and tested.
That creates a problem for institutions that depend on moral authority. A single statement may be explained away; however , a recurring pattern is much harder to set aside. When an international official refers to one conflict more often than others, adopts harsher language for one actor, or returns repeatedly to one set of violations while giving less attention to comparable conduct elsewhere, the cumulative effect becomes political, and shapes what the world sees.
Edward Bernays understood almost a century ago that public opinion does not form spontaneously. In Propaganda, he described how an “invisible government” of elites could use organized persuasion to shape public attitudes in politics, business, and culture. His insight was unsettling because it treated opinion less as a natural expression of the public mind than as something guided through repetition, authority, symbols, headlines, institutions, and social pressure.
Modern propaganda rarely announces itself. It operates through managed emphasis and tells the public where to look, what to feel, and how to rank moral urgency. It does this through repetition rather than open command, through framing rather than force.
International institutions now play a central role in that process. Their statements travel through newsrooms, advocacy networks, diplomatic channels, social media, and university campuses. A phrase used by a secretary-general can become a headline. That headline becomes a slogan and that slogan becomes the moral frame through which later facts are interpreted.
That is why the JPPI study offers a useful starting point. The issue reaches beyond one official or one institution as the study exposes a lacuna in how institutional bias is usually discussed. The most consequential bias may lie less in a single sentence than in accumulated patterns of attention. Repetition creates priority. Tone creates judgment. Omission creates hierarchy.
The media then amplifies the frame as news organizations often treat institutional language as unchallenged neutral authority. When the UN, a major NGO, or an international court uses a formulation, journalists repeat it, often with little scrutiny. The repetition confers legitimacy with the original statement becoming background assumption. Readers rarely see the chain by which a contested judgment becomes common language.
Matti Friedman has written powerfully about this process from inside the media world. His central insight is simple: news coverage does not merely report reality; it selects, arranges, and assigns meaning. In the Israel-Hamas conflict, that selection has often produced a familiar script: Israel as the active force, Palestinians as passive sufferers, and militant actors as background conditions. Once that frame takes hold, later reporting tends to fit inside it.
Media-monitoring research has made similar claims through data. Patterns in headlines, source selection, imagery, casualty emphasis, and story placement can tilt public perception before readers reach the body of an article. A headline can assign agency. A carefully selected photograph can establish sympathy. A repeated formulation can make a political interpretation feel like description.
This is where institutional language and media amplification meet. An international organization supplies the moral vocabulary. News outlets distribute it. Activists intensify it. Politicians respond to it. Eventually the frame begins to influence policy: diplomatic pressure rises, legal scrutiny narrows, and public opinion shifts.
The result is an impaired field of perception. Later facts still matter, but they enter an environment already shaped by earlier repetition. Corrections travel more slowly than accusations. Context arrives after outrage. Complexity struggles to compete with a phrase that has already hardened into conventional wisdom.
Artificial intelligence did not create this problem; it makes the architecture beneath the pattern visible. AI tools can scan thousands of statements, classify sentiment, track repetition, compare references across countries or conflicts, and detect asymmetries of tone. AI gives form to what was once intuition. It pulls back the curtain on patterns that worked in shadow and exposes the machinery of institutional language.
There is little effective interdiction between institutional framing and media amplification. Once a narrative leaves an international podium, it moves rapidly through systems that reward repetition over verification. Each retelling makes the original frame more familiar. Familiarity then begins to masquerade as truth.
Critics may inveigh against bias, but accusation alone rarely changes anything. The stronger response is measurement. Institutions that claim neutrality should be studied as rigorously as governments, corporations, and political campaigns. Their statements should be compared across conflicts. Their language should be analyzed over time. Their patterns of emphasis should be made public.
This does not require censorship or ideological policing. It requires transparency. International organizations should publish searchable archives of official statements, disclose criteria for public emphasis, and submit communications to independent review. Media organizations should identify when legal or moral claims originate from contested institutional sources. Scholars and civic groups should build public dashboards tracking frequency, tone, and omission across major conflicts.
Such tools would not settle every dispute. They would make evasion harder. A system that measures institutional language can reveal when one conflict receives disproportionate attention, when one actor receives unusually harsh treatment, and when comparable conduct elsewhere receives softer language or less attention.
Language is power invisibly shaping perception. International institutions shape reality long before resolutions pass or sanctions appear. They determine which suffering becomes central, which violations become urgent, which actors become villains, and which contexts disappear from view. Their authority lies partly in law, partly in legitimacy, and partly in repetition.
Bernays saw that modern societies could be guided by organized persuasion. Our era has expanded the machinery. The UN, NGOs, media platforms, algorithms, universities, and advocacy groups now participate in a vast system of narrative production. The system does not need a central command. It only needs repeated cues moving in the same direction.
AI analysis gives the public a way to examine that system. It can show whether institutional language behaves neutrally or creates cumulative pressure against selected targets. It can move the debate from feeling to evidence.
The solution begins with measurement, but it cannot end there. Institutions that speak with moral authority must accept scrutiny of their own language. Media outlets must stop treating institutional phrasing as inherently neutral. Democracies must challenge narrative patterns early, before repetition turns them into settled assumptions.
Political reality is made not only by tanks, treaties, or votes. It is also made by words repeated until they become the atmosphere. Measuring that atmosphere is now possible. The harder task is demanding accountability from those who create it.
Aaron J. Shuster is an award-winning cinematist, writer, and essayist who examines history, strategic culture, and the civilizational pressures shaping Western civilization.
