A civic community needs more than shared values. It needs enough shared reference points for people to discover disagreement about the same events, laws, needs, and persons. Citizens may interpret a budget, a war, or a school closure differently, but they must still be able to ask what happened, which sources are accountable, and what evidence could change a judgment. Without some common world, disagreement stops being a contest over reality and becomes a contest between sealed realities.
Artificial intelligence can intensify this problem through personalized synthetic feeds. A system can generate an apparently endless stream of explanations, images, voices, summaries, and arguments adapted to one person’s fears and preferences. Another person receives a different stream, equally fluent and emotionally plausible. Strategic actors can flood a public conversation with conflicting material until citizens cannot tell whether they are seeing a coordinated campaign, ordinary error, or an event at all.
This essay is about the civic conditions for a common world. It is not an argument for a hive mind, unanimous opinion, or one centralized consciousness. Nor is it a replacement for the foundations of verification: identifying sources, checking evidence, and distinguishing truth from consensus remain necessary. The question here is what communities and institutions must do after those foundations are recognized, so that verified claims can still become shared reference points for communication and solidarity.
From a Shared Feed to Separate Realities
Personalization is not automatically a civic evil. Translation, accessibility tools, local information, and explanations fitted to a learner can help people participate. No society has ever given every citizen exactly the same knowledge. Families, professions, languages, neighborhoods, and experiences rightly shape what people notice.
The danger arises when personalization governs not only presentation but the underlying set of facts a person can encounter. If a platform optimizes for retention, it can learn that outrage, humiliation, or fear keeps a user engaged. Generative systems then make it cheap to supply an individually tailored rationale for any emotional conclusion. One citizen sees a fabricated expert explaining a conspiracy; another receives a persuasive denial; a third is shown nothing because the subject does not fit a profitable prediction. Each feed feels complete while omitting the possibility that others are inhabiting a different informational environment.
Synthetic abundance also changes the economics of attention. A false claim no longer needs to persuade everyone. It can be aimed at a narrow group, varied across thousands of accounts, and replaced before journalists or public bodies can document it. Automated replies can manufacture the impression of a crowd. Conflicting versions can be released deliberately, not to establish one lie but to make inquiry seem pointless. The success condition is sometimes not belief. It is exhaustion.
Strategic Flooding and Civic Fragmentation
Strategic flooding exploits a basic weakness in public reasoning: people have limited time to compare sources, while a network can generate claims without comparable cost. When every event produces thousands of plausible accounts, citizens may retreat to identity, loyalty, or the account that arrived first. Institutions are then pressured to respond to volume rather than evidence. A false allegation can remain socially effective even after it is disproved because the original impression has already shaped a vote, a donation, or a threat.
Fragmentation does not require that people disagree about everything. It is enough for them to lose confidence that disagreement can be resolved through shared inquiry. If one group treats a public health record as a trustworthy source and another regards it as fabricated, they do not merely have competing policy preferences. They lack a common procedure for finding out what the record says. The result is a weakened capacity to mourn together, deliberate together, and accept an outcome without regarding every institution as an enemy.
The costs fall most heavily on people who need cooperation. A disaster response requires common maps and warnings. A courtroom requires facts that opposing parties can challenge within a shared procedure. Neighbors need reliable information about a missing person or a contaminated water supply. Solidarity is difficult when each appeal for help can be answered by an individually generated story that discredits the person asking.
A Catholic Account of the Common World
Catholic social teaching does not define the common good as everyone having the same opinions. In Gaudium et Spes 26, the common good concerns the social conditions that allow persons and communities to pursue their fulfillment, and it is inseparable from the dignity of each person. Those conditions include truthful communication and the possibility of participation. A population that cannot identify trustworthy facts cannot effectively deliberate about the responsibilities it shares.
The tradition also resists two opposite errors. One is atomism: the idea that each person can live from a private information world without obligations to others. The other is coercive uniformity: the idea that social peace requires one authorized interpretation. Fratelli Tutti 45–50 warns about forms of digital communication that can encourage isolation and aggression, while its emphasis on social friendship points toward patient encounter rather than manufactured unanimity. A common world is a shared field for truth-seeking and encounter, not a command that consciences must reach identical conclusions.
Subsidiarity is relevant here as well. Families, schools, local journalists, professional associations, and civic groups possess knowledge that a distant platform cannot replace. But subsidiarity is not a license for every small community to become an information island. Solidarity requires institutions capable of sharing reliable findings across boundaries, especially when harms cross a neighborhood, diocese, or nation. The moral task is to distribute responsibility without dissolving the references on which cooperation depends.
Communication Is More Than Delivery
A personalized answer can be factually correct and still weaken public communication if no one knows what others were shown. Civic speech needs a layer of common visibility: public documents, stable definitions, accessible records, and identifiable accounts of who is responsible. People can then receive explanations suited to their language or ability while retaining access to the same underlying proposition.
This distinction matters for public institutions. A government should not announce a policy through millions of individually optimized messages whose differences conceal the actual commitment. A school should publish a stable rule before offering a chatbot to explain it. A parish can use different formats for a pastoral message while keeping a durable, reviewable version available to everyone. Personalization should improve access to a common statement, not replace the statement with private persuasion.
The Human Work of Listening
Synthetic systems are good at producing responses and poor at bearing the moral cost of being in relationship. Communication requires listening, the willingness to be corrected, and the recognition that a neighbor is not merely an audience segment. The more a feed predicts what will satisfy a person, the more important it becomes to preserve places where people encounter an unoptimized account: a town meeting, a classroom discussion, a newsroom, a jury process, a family conversation, or a parish gathering.
These encounters are not valuable because face-to-face life is always truthful. Humans manipulate one another in person too. They are valuable because accountable presence makes questions and repair possible. A person can ask for a source, notice a wounded neighbor, clarify a misunderstanding, and accept responsibility. Technology should widen this capacity for encounter rather than make each citizen a target of an invisible persuasion system.
Institutions That Preserve Shared Reference Points
First, public institutions and serious organizations should maintain canonical records. Meeting minutes, regulations, budgets, election results, scientific data, and emergency notices should be published in stable, accessible forms with dates, responsible offices, revision histories, and contact points. Summaries generated for different audiences should link back to the same underlying record. A common reference point does not have to be a single website, but it must be findable and accountable.
Second, platforms should expose the basic conditions of distribution. Users need to know when content is synthetic or substantially altered, who paid for political communication, whether an account is automated, and why a message is being recommended when that information can be provided without enabling abuse. Labels will not defeat every deception, especially when material is reposted without its metadata. They still help establish norms and give researchers and users a way to compare the feed with the source.
Third, institutions should support trusted intermediaries. Local journalism, libraries, schools, public-interest archives, fact-checking organizations, and professional bodies can test claims and translate them into forms communities can use. Their independence matters: government communication should not become a requirement that citizens accept an official narrative. Funding, governance, corrections, and transparent methods are part of the trustworthiness of an intermediary.
Fourth, communities need procedures for crisis communication. An emergency channel should state what is known, what is uncertain, when it will be updated, and where corrections will appear. Repetition of a false rumor can amplify it, so officials should answer the underlying need for information without unnecessarily reproducing harmful content. The goal is not to win a speed contest with synthetic media but to make a reliable path easier to follow than the flood.
Common Reference Points, Not Common Conclusions
The target is narrower than a common feed or a common conclusion. A common reference point is a source that people can locate, inspect, challenge, and revisit: the same scope, date, revision history, and uncertainty should remain beneath translations, summaries, audio descriptions, and other personalized explanations. Citizens may still judge the source differently. The obligation is to make disagreement answerable to the same publicly contestable object rather than to force agreement with whoever maintains the record.
Canonical does not mean infallible or final. Every record should name its custodian, expose a correction route, preserve minority testimony that contests it, and show when its scope or wording changed. A counter-public can be a necessary route to visibility when an official archive excludes a group; it should be able to link its testimony to the shared record without surrendering its independence. A process that cannot be challenged is not a common reference point but a gatekeeper.
The causal mechanisms also need separation. Tailored translation or accessible explanation may widen participation; synthetic flooding may exhaust attention; chronological presentation may change little when partisan identity dominates. Labels or a common source are therefore interventions to test, not guarantees of convergence. Cross-group audits should measure factual convergence, correction uptake, emergency coordination, source diversity, accessibility, false-label rates, correction speed, and whether minority testimony becomes more or less visible. No such field evidence is supplied here, so the argument is a normative design criterion and a research agenda, not a claim that canonical records will reliably change beliefs.
The Objection: Is a Common Reality Impossible?
The strongest objection is that there has never been one common reality. Newspapers disagreed, authorities lied, minorities were excluded from official archives, and “shared facts” have sometimes meant the powerful forcing their account on everyone else. Calls for a common information space can therefore sound like nostalgia or a plan for censorship. Personal feeds may give marginalized people sources and communities that gatekeepers once denied them.
This objection is right to reject a romantic past. A common world must be more inclusive and corrigible than older public spheres. It should preserve dissenting testimony, disclose uncertainty, and allow independent institutions to challenge official records. The answer to exclusion is not to abolish common reference points; it is to make the process of constructing them more open, plural, and accountable.
Another objection is that provenance, labels, and platform governance can be evaded. They can. No technical measure makes a network truthful. The realistic standard is not perfect prevention but increased accountability and reduced reach for deliberate flooding. A public record, a visible correction, and a known responsible editor give citizens more to work with than an untraceable stream of generated assertions. Institutional redundancy matters: when one source fails, others can compare and correct it.
Finally, some fear that slowing personalization will sacrifice accessibility. It need not. A blind reader can receive an audio description; a newcomer can receive a translation; a student can ask for a simpler explanation. The safeguard is to preserve the same source, scope, and revision status beneath those adaptations. Inclusion and commonality are allies when personalization is treated as a bridge to shared reality rather than a substitute for it.
Practical Habits for Citizens and Leaders
Citizens can diversify their sources without pretending every source is equally credible. They can save the original notice rather than relying on a screenshot, ask what would count as disconfirming evidence, and resist forwarding material whose main effect is panic or contempt. They can support local reporting and organizations that show their methods. In conversation, they can begin by identifying the proposition actually in dispute instead of treating an entire group as irrational.
Leaders have a higher duty because their statements organize attention. They should publish corrections plainly, avoid using generated testimonials or undisclosed synthetic voices, and distinguish verified information from interpretation. A leader who exploits confusion may gain a short-term advantage while damaging the shared conditions that make leadership possible. In Catholic terms, authority is ordered to service and the common good, not merely to control of the narrative.
Sources and qualifications
The Catholic account draws on Gaudium et Spes 26; Fratelli Tutti 45–50, 165–167, and 199–206; theCatechism of the Catholic Church 1877–1889 and 1905–1912; and the Pontifical Council for Social Communications, Ethics in Communications (2000). These sources support concern for truthful communication, human dignity, solidarity, subsidiarity, social friendship, and the common good. They do not prescribe a particular platform design or settle every policy dispute.
The technical claims concern recommendation systems, generative media, coordinated inauthentic activity, and provenance initiatives such as C2PA. Labels, watermarking, ranking changes, and transparency reports have limitations and can be evaded; no one measure guarantees a common world. The proposals are institutional principles rather than a promise of perfect moderation. They should be tested for effects on privacy, accessibility, free expression, and marginalized communities.
References to civic institutions and media practice are general, not legal advice. Election, privacy, platform, defamation, and public records law vary by jurisdiction. A common reference point must not become an excuse to suppress lawful dissent or compel agreement. The narrower claim is that democratic and communal cooperation requires accessible sources, accountable communication, visible correction, and opportunities for people with different judgments to investigate the same world.
Concluding synthesis
AI can make every person feel unusually well informed while making a society less able to know together. Personalized synthetic feeds, strategic flooding, and automated persuasion exploit the distance between private plausibility and public accountability. If citizens lose the ability to identify common reference points, solidarity becomes a slogan and communication becomes targeting by another name.
The remedy is not a hive mind and not an impossible return to a single shared feed. It is a plural public world: truthful sources that can be challenged, records that remain accessible, intermediaries that explain their work, technologies that disclose their role, and human places where listening can lead to correction. Verification remains the foundation, but institutions must carry verified knowledge into common life. A society preserves its reality when it gives people both freedom to judge and enough shared ground to judge with one another.
Source Notes
The Council’s account of the common good and the social conditions required for persons and communities to flourish.
A Catholic account of social friendship, encounter, digital isolation, and solidarity across difference.
The Church’s teaching on social nature, the common good, solidarity, subsidiarity, and truthful communication.
A Catholic framework for truth, responsibility, media power, participation, and communication ordered toward the common good.
A provenance initiative relevant to labeling and tracing synthetic or transformed media, with acknowledged limits against coordinated deception.
International principles for human agency, transparency, accountability, diversity, and the social effects of AI-mediated information.
Editorial revision note
This essay completed one self-editing, cross-essay consistency, and strongest-criticism pass. Its controlling thesis was preserved while objections, qualifications, evidence limits, and practical consequences were strengthened.