The Crowd Is Scripted: How Bot Networks and Coordinated Accounts Manufacture a Fake Consensus

The illusion of “everybody thinks so”

Imagine scrolling through social media and encountering a controversial claim. The post has thousands of likes. Hundreds of people are repeating the same argument. Comment after comment insists that the truth is obvious and that anyone who disagrees is either stupid or dishonest.

You begin to wonder: Am I missing something? Can so many people really be wrong?

That reaction is precisely what makes manufactured consensus so useful to anyone trying to manipulate public opinion.

Social media allows people to express genuine agreement, organize around shared beliefs and challenge powerful institutions. But the same platforms can also be exploited by networks of fake accounts, automated bots, paid operators and coordinated groups designed to make a message appear more widely supported than it really is.

The objective isn't always to persuade you with better evidence. Sometimes, it's to persuade you that the argument has already been settled by everyone else.

The deception isn't necessarily the message. It's the false impression of who supports it.

How a manufactured crowd works

Coordinated manipulation can take several forms. Not every campaign uses all of them, and not every account involved is automated or fake.

1. Bots create activity at scale

A bot is an account that uses automation to perform some or all of its online activity. Bots can serve legitimate purposes, such as distributing weather alerts or providing updates. The problem arises when automated accounts are used deceptively to amplify claims, flood discussions or create misleading impressions of popularity.

A coordinated network can share the same links, repeat identical talking points, reply to prominent accounts or rapidly boost a particular post.

To an ordinary user, the resulting activity may look like widespread spontaneous interest. The accounts may be numerous, but the apparent crowd can be much less independent than it seems.

2. Fake accounts impersonate ordinary people

A network doesn't need to announce itself as a coordinated campaign. Its accounts may present themselves as local residents, concerned parents, enthusiastic customers, independent commentators or ordinary citizens.

Some may have profile photographs, personal biographies and years of apparent activity. Others may borrow names, images or details from real people.

The point is to make a centrally directed message look as though it is emerging independently from many different people.

Meta has documented cases in which coordinated networks used fake accounts to mislead people about who was behind an operation and what it was trying to do.

3. Real people can be recruited into the performance

Not every coordinated campaign relies on bots. Some use paid commenters, organized supporters, marketing contractors or groups of genuine people following a shared plan.

That distinction matters. A coordinated campaign can include real humans and still mislead its audience if it conceals who is organizing the activity or falsely presents a managed campaign as spontaneous public opinion.

Organized advocacy is not automatically deceptive. People are entitled to campaign together. The warning sign is concealed coordination combined with a misleading impression of independent support.

4. One message appears to come from everywhere

Imagine dozens of accounts posting the same unusual phrase, sharing the same link within minutes or repeating an identical talking point across several groups.

Each post may look like another person independently reaching the same conclusion. But if the accounts are centrally coordinated, the apparent agreement may be an echo of one operation rather than a collection of independent judgments.

This is sometimes called astroturfing: the creation of an artificial grassroots appearance for a campaign that has an organized sponsor or hidden source.

The technique can be used for political influence, commercial promotion, reputation management or other efforts to shape public perception.

Why a fake consensus can be so persuasive

Manufactured consensus exploits an ordinary feature of human psychology: we pay attention to what other people appear to believe.

When we lack time, expertise or reliable information, other people's reactions can help us judge what matters. If a product has thousands of positive reviews, we may assume it's popular. If a claim is repeated by seemingly independent people, we may assume it has been verified. If a viewpoint appears dominant, we may hesitate to challenge it.

These shortcuts are not signs of stupidity. They're part of how people navigate a complicated world.

But they can be exploited.

Popularity starts to look like proof

A claim that attracts enormous engagement can feel credible simply because it is everywhere. Yet engagement tells us that content is receiving attention—not whether its claims are accurate.

Even genuine popularity doesn't establish truth. A widely shared falsehood remains false, and a little-known fact doesn't become untrue because few people discuss it.

Repetition creates familiarity

When we encounter the same assertion repeatedly, it can begin to feel familiar and easier to process. That feeling of familiarity may be mistaken for evidence that the claim is reliable.

A coordinated campaign can take advantage of this effect by repeating a message across accounts, groups and platforms.

Seeing the same claim ten times, however, is not the same as seeing ten independent sources verify it.

Disagreement can appear pointless or dangerous

A manufactured majority can make people who disagree feel isolated. They may stay silent, avoid asking questions or assume that their own doubts are unreasonable.

The result can be a distorted picture of public opinion: people who appear to agree publicly may not all agree privately, and people with different views may become less willing to express them.

This doesn't mean every popular viewpoint is manufactured or that every quiet person secretly disagrees. It means visible online activity is not a reliable census of what everyone thinks.

The algorithmic amplifier

Social platforms use different ranking and recommendation systems to decide which posts users see. Engagement can be one of the signals these systems consider, although the precise signals and their importance vary by platform and change over time.

That creates a potential opportunity for manipulation.

If coordinated accounts generate artificial activity around a post, they may help make the content look more engaging or draw additional attention to it. If the platform's systems respond to those signals, the content may gain further visibility.

Real users can then encounter the post, react to it and share it themselves. The original artificial activity may have helped attract genuine attention, making it increasingly difficult to distinguish the initial push from the later response.

This is not a guarantee that bots can make any post go viral. Platforms use different detection and ranking systems, and artificial engagement doesn't automatically translate into widespread reach.

But the underlying vulnerability is clear: when attention becomes a signal of importance, manipulating attention can become a way to manipulate visibility.

The crowd doesn't have to be entirely fake to create a misleading impression. A relatively small coordinated group may help shape what a much larger genuine audience sees.

Who benefits from manufacturing consensus?

The tactic is not confined to one political party, country, ideology or industry. Its potential beneficiaries include several kinds of actors.

• Political operators may try to make a candidate, policy or political narrative appear more popular or widely accepted than it is.
• Commercial interests may use deceptive engagement to make a product, service or brand seem more popular or trustworthy.
• Reputation managers may try to drown out criticism or create the appearance of widespread praise.
• Influence operations may seek to intensify existing divisions, undermine trust or make competing groups believe they are under attack.
• Scammers and dishonest promoters may exploit artificial social proof to make dubious offers or misleading claims look credible.

The motives differ, but the basic mechanism is similar: manipulate the apparent source, scale or independence of support.

There is an important distinction here. A campaign can promote a viewpoint that is true, false or somewhere in between. The existence of coordinated manipulation does not, by itself, prove that the message is false.

The deception lies in misrepresenting the activity behind the message.

How to recognize a suspicious pattern

There is no foolproof way to identify a bot or coordinated network just by looking at a profile. Some automated accounts are legitimate, some genuine people behave repetitively, and sophisticated fake accounts can look convincing.

Instead of relying on a single clue, look for patterns.

Watch for these warning signs:

• Sudden, unusual bursts of activity. A claim attracts a wave of nearly identical posts or replies in a short period.
• Repeated wording. Numerous accounts use the same unusual phrases, slogans, links or talking points.
• Suspiciously synchronized behavior. Accounts repeatedly post, share or react in close coordination.
• Questionable identities. Profiles have inconsistent details, stolen-looking images, implausible histories or signs of impersonation.
• A mismatch between activity and credibility. An account generates enormous engagement but offers little verifiable information about who runs it or where its claims originate.
• Pressure to conform. Replies insist that “everyone knows” something is true, attack anyone who questions it or use apparent majority support as a substitute for evidence.

None of these clues proves manipulation on its own. A breaking-news event can produce a sudden burst of genuine activity. Activists may legitimately share common slogans. Real people can repeat information they trust.

The question is whether multiple clues point toward a coordinated pattern—and whether there is independent evidence to support that conclusion.

Six steps to investigate a suspicious online crowd

You don't need specialist software or an investigative team to apply some basic checks.

1. Separate the claim from the crowd.

Ask what the post actually asserts. If you ignore the likes, comments and follower counts, what evidence remains?

2. Look for independent confirmation.

Search for reliable sources that verify the underlying claim. Seek original documents, direct statements, credible reporting or relevant research—not merely more accounts repeating the same assertion.

3. Compare the accounts' behavior.

Look for repeated wording, synchronized posting, identical links or other unusual similarities. Treat these as clues, not proof.

4. Check where the information originated.

Follow the claim back toward its earliest identifiable source. Does it lead to a named person, an original document or verifiable reporting? Or does it circulate in a closed loop of accounts citing one another?

5. Look for platform investigations.

When available, consult credible reporting or official platform announcements documenting coordinated networks. These findings can provide stronger evidence than trying to diagnose individual accounts from profile pages.

6. Don't confuse uncertainty with proof of a conspiracy.

You may be unable to establish whether a crowd is genuine. That's a reason to withhold judgment, not a reason to assume that every popular view is manufactured.

The goal is not to become suspicious of everyone. It's to stop treating visible agreement as a substitute for independent evidence.

What the numbers can't tell you

A post with 100,000 likes may reflect genuine enthusiasm, coordinated promotion, paid engagement or some combination of factors. The number alone cannot tell you which.

A video with millions of views doesn't prove that millions of people believe its message. A trending topic doesn't establish that it represents the majority view. A flood of comments doesn't reveal how many commenters are independent, how many have seen the same source, or how many agree with the underlying claim.

Even a large, authentic crowd can be mistaken.

And a small minority can sometimes be right.

The more consequential the claim, the more important it is to ask what supports it independently of its popularity.

The deeper deception: confusing visibility with reality

Social media shows us a stream of selected, ranked and amplified activity. It does not give us a transparent view of society.

We see the posts that reach us, the comments that remain visible, the accounts that attract attention and the opinions that appear most frequently in our feeds. We rarely see the full population behind those signals—or the activity that went undetected, was removed or never reached us.

That gap creates room for manipulation.

A coordinated network may not need to convince you that a particular claim is true. It may only need to make you believe that everyone else already accepts it.

Once you think you're in the minority, you may become less willing to question the message. You may share it to stay on the winning side, or remain silent to avoid ridicule. Other people may react in the same way, making the apparent consensus seem even stronger.

A manufactured impression can therefore influence genuine behavior, even when the original crowd was partly or largely artificial.

The antidote is not cynicism. It's intellectual independence: the willingness to examine a claim on its merits, regardless of how loudly or how often it is repeated.

Evidence: STRONG — Research and documented platform investigations establish that coordinated inauthentic networks and automated accounts have been used to manipulate online visibility and public discussion. The presence and influence of any specific network must be established with case-specific evidence.