When Your Algorithm Becomes Your Evidence
How social media can turn repeated exposure into certainty, why trusting your gut isn't always enough, and what psychology reveals about the danger of mistaking conviction for truth.
Have You Ever Been Absolutely Certain, Only to Discover You Were Wrong?
Think about a time when you were completely convinced of something.
Maybe it was a person you fell in love with. You were certain they were the one. You felt it in your gut. You saw a future together, and no amount of advice from friends or family could convince you otherwise.
Then, months or years later, you discovered things you hadn't seen before. Perhaps the relationship wasn't what you thought it was. Maybe the warning signs had always been there, but you were so convinced of what you believed that you explained them away.
Or perhaps it was a business opportunity, an investment, or a decision you felt absolutely certain would work out.
And then it didn't.
Looking back, you might wonder how you could have been so sure.
The answer is surprisingly simple: Feeling certain and being correct are two very different things.
Our brains are remarkably capable of creating confidence around incomplete or inaccurate information. We do it in relationships, financial decisions, professional judgments, and countless everyday situations.
So why would our beliefs about conspiracy theories be any different?
And what happens when the source continually reinforcing those beliefs is an algorithm designed to keep us watching?
Your Gut Is Not a Fact-Checking System
We often hear people say they can just feel when something is true.
They trust their intuition. Something resonates. Pieces appear to fit together. Suddenly, an explanation that once seemed impossible feels obvious.
Intuition can be valuable. It helps us recognize patterns, respond quickly, and make decisions based on experience.
But intuition isn't infallible.
Here's a question we rarely ask ourselves: How accurate has your gut actually been?
Look back over your life. Think about the people you were certain you could trust, the relationships you were convinced would last, the decisions you knew were right, or the situations where you were absolutely sure you understood what was happening.
How many times did you later discover that your intuition had been wrong?
For many of us, the answer is more often than we'd like to admit.
Yet somehow, when we encounter an extraordinary claim online and something inside us says, This resonates. I can feel that it's true, we treat that feeling as though it has suddenly become a reliable source of evidence.
If your gut has been wrong before, perhaps repeatedly, why should its reaction to a YouTube video be considered proof of anything?
That doesn't mean intuition is worthless. It means intuition deserves the same scrutiny as any other source of information, especially when its track record includes mistakes.
Sometimes what feels like recognition is familiarity. Sometimes what feels like evidence is coincidence. And sometimes what feels like discovering the truth is simply discovering a story that fits what we already suspected.
Psychologists have studied the relationship between intuitive thinking and conspiracy beliefs.
A 2022 research review published in Current Opinion in Psychology found that conspiracy belief is often associated with greater reliance on intuition and less reflective thinking. The authors also acknowledged limitations in establishing the underlying causes.
That distinction matters. The research doesn't mean intuitive people are inherently irrational or incapable of critical thought. It means that intuition alone is not an adequate method for determining whether an extraordinary claim is true.
Consider something as simple as hearing a strange noise in your house at night.
You might immediately feel that someone is inside.
Your heart races. Your mind starts assembling possibilities. Every little sound becomes more significant.
But when you investigate, you discover an open window rattling in the wind.
The fear was real. The conviction was real.
The explanation was wrong.
The intensity of a feeling tells us something about our emotional response. It does not establish the accuracy of our conclusion.
When Seeing Something Repeatedly Starts to Feel Like Proof
Imagine watching a video claiming that a famous celebrity secretly died years ago and was replaced by another person.
The creator shows side-by-side photographs. They point to facial similarities, unusual gestures, and supposed inconsistencies in public records.
At first, you're skeptical.
Then another video appears. Different creator, same theory.
Then another.
Soon, you're seeing dozens of examples. The comments are filled with people agreeing, adding observations, and congratulating one another for noticing something supposedly hidden from the public.
Eventually, you might begin thinking: There must be something to this. Why else would so many people be talking about it?
But consider what has actually happened.
You haven't necessarily discovered new evidence.
You've encountered the same basic claim repeatedly, sometimes presented by different people who may all be relying on the same original source.
Psychologists call this the illusory truth effect.
The principle is that repeated exposure to a statement can increase how truthful it seems, even when the statement is false.
A 2026 meta-analysis published in Nature Communications examined 182 studies involving more than 31,000 participants. Researchers found a small but consistent effect of repetition on perceived truth, although its strength varied across situations.
Repetition doesn't automatically convince everyone. Yet familiarity can make claims easier to process, and that ease can influence our judgments.
Something we've heard twenty times may begin to feel more credible than something we're hearing for the first time, even when the unfamiliar explanation is supported by substantially stronger evidence.
Familiarity is not verification. Repetition is not independent confirmation.
The Algorithm Doesn't Know You've Found the Truth
Imagine researching Flat Earth.
You watch one video. Then another. You finish a long interview with someone who claims to have uncovered evidence that NASA has concealed the planet's true shape.
Soon, your recommendations begin including more Flat Earth discussions, alternative cosmology, alleged government cover-ups, and related conspiracy content.
It may start to feel as though evidence is everywhere.
But an important distinction has been lost.
Your recommendation feed is not conducting an independent investigation into the shape of the Earth.
It is selecting content based on signals such as viewing behavior, interests, and predicted engagement.
A 2023 study published in Proceedings of the National Academy of Sciences audited YouTube's recommendation system using 100,000 simulated users. Researchers found that the system often recommended ideologically compatible content and that some users encountered increasing proportions of material from problematic channels, including conspiracy-related sources.
However, the researchers did not find that recommendations inevitably became more ideologically extreme as viewing continued.
That nuance is important.
Algorithms do not control everyone's beliefs, and users are not powerless. People choose what to watch, search for, believe, reject, and share.
But recommendation systems can influence the information environment in which those decisions occur.
And that creates an unusual problem.
The more interested you become in a claim, the more frequently you may encounter material reinforcing that interest. The increasing volume can feel like mounting evidence, even when it isn't.
Ten videos repeating one unsupported or deliberately false theory do not provide ten independent pieces of evidence.
They may simply represent ten people repeating the same mistake... or deception.
Why Contradictory Evidence Can Become Part of the Conspiracy
Here is where conspiracy beliefs can become especially difficult to evaluate.
Imagine someone believes that powerful individuals secretly control world events.
When confronted with evidence contradicting the theory, they may respond that the evidence itself has been manipulated by those powerful individuals.
A scientific institution disagrees? It's compromised.
A historian presents original documents? The documents were fabricated.
A journalist investigates the claims? The media is protecting the conspiracy.
A former believer changes their mind? They've been threatened or paid to remain silent.
Notice what happens.
The theory no longer needs to explain contradictory evidence on its merits. It can absorb the contradiction and use it as additional support.
This is sometimes called a self-sealing belief, although not every conspiracy theory or believer operates this way.
The danger appears when a person develops a standard in which information supporting the belief counts as evidence, while information contradicting it automatically counts as evidence of a cover-up.
Under those conditions, what could ever establish that the original belief was mistaken?
A theory that cannot possibly be wrong has lost one of the essential qualities that makes a claim testable.
This does not mean genuine conspiracies never occur. History contains well-documented examples of deception, secrecy, and institutional wrongdoing.
The difference is that genuine investigations require evidence capable of distinguishing a true allegation from a false one.
Suspicion can justify investigation.
It cannot replace the investigation.
When Being Right Becomes Part of Who You Are
There is another psychological dimension to consider.
Sometimes beliefs become connected to identity, belonging, and self-worth.
A person may initially encounter a conspiracy theory out of curiosity. Over time, they begin following creators, joining groups, attending online discussions, and developing friendships around a shared worldview.
Perhaps they've spent years telling others that they understand truths most people refuse to see.
Now imagine discovering compelling evidence that a foundational belief is mistaken.
Changing their mind might mean more than correcting a factual misunderstanding.
It could mean reconsidering years of conversations, arguments, friendships, and personal decisions.
A 2025 systematic review and meta-analysis in Psychological Bulletin examined 279 independent studies involving more than 137,000 participants. The researchers found associations between conspiracy beliefs and motivations involving understanding the world, managing uncertainty or threat, and social identity.
These associations were generally modest and varied across studies. They don't describe every individual believer.
But they help explain why a factual disagreement can sometimes feel like a personal attack.
If a belief has become part of someone's identity, challenging it may feel like challenging the person themselves.
And if an online community continually rewards certainty while discouraging doubt, reconsidering a belief may become socially uncomfortable.
The longer a belief becomes intertwined with who we are, the harder it can become to separate the evidence from the emotional cost of being wrong.
Outrage Can Make a Story Travel Further Than Evidence
Consider two videos.
One presents a careful, evidence-based explanation of why photographs, measurements, and observations support Earth's approximately spherical shape.
The other announces that scientists, governments, teachers, and entire generations have been deliberately deceiving humanity.
Which video is more likely to generate an emotional reaction?
The second offers something the first doesn't: betrayal, secrecy, a powerful enemy, and the possibility of possessing forbidden knowledge.
Research suggests these emotional dynamics matter.
A 2024 study published in Science examined more than one million Facebook links, tens of thousands of posts on Twitter, and additional behavioral experiments.
Researchers found that misinformation sources tended to evoke more outrage than trustworthy sources, and that outrage made people more willing to share misinformation, sometimes without reading it first.
This matters because online attention does not require agreement.
A person passionately defending a conspiracy theory may engage with the content.
Someone ridiculing the same theory may also engage.
Both reactions can contribute to visibility, depending on how the platform distributes content.
For creators who receive advertising revenue, sponsorships, donations, or paid memberships, attention may also have financial value.
That doesn't mean every controversial creator is knowingly dishonest. Some genuinely believe their claims.
But it does mean a creator's financial incentives and a viewer's search for truth are not necessarily aligned.
Popularity tells us that people are paying attention.
It doesn't tell us whether the information deserves their belief.
Why Real Life Can Correct Us, but Our Feeds May Not
Return to the example of a relationship you were once certain about.
You believed you'd found the perfect partner.
Perhaps you ignored warnings, explained away inconsistencies, and focused on the experiences supporting your expectations.
But eventually, ordinary life introduced information you couldn't avoid.
You experienced the relationship over time. You witnessed behavior, encountered contradictions, and had opportunities to reconsider your original judgment.
You might still resist those observations, but the relationship existed independently of your expectations.
Social media can operate differently.
If a conspiracy claim is reinforced primarily through online videos, posts, and communities, the person may spend years encountering interpretations that support it without seriously examining the original evidence.
And unlike a relationship or business investment, certain extraordinary claims may never produce an obvious personal consequence that forces reconsideration.
Someone can believe for decades that a celebrity secretly changed identities, that an ancient civilization possessed impossible technology, or that world leaders are shape-shifting beings.
Everyday life may continue without directly testing those beliefs.
Meanwhile, the algorithm keeps serving related material.
Of course, contradictory evidence exists outside the feed, and sometimes within it. Nothing prevents an interested viewer from seeking it.
But if a person has learned to dismiss all opposing information as deception, the opportunity to correct a mistaken belief becomes increasingly limited.
The real danger isn't merely believing something false. It's losing the ability to recognize what would demonstrate that it's false.
The Difference Between Questioning Everything and Believing Anything
Skepticism is valuable.
Science depends on questioning explanations, challenging assumptions, and testing existing knowledge.
Investigative journalism depends on refusing to accept every official statement without scrutiny.
And ordinary people should absolutely feel comfortable asking difficult questions.
But questioning established explanations does not automatically make an alternative explanation credible.
Rejecting one claim is not evidence that another claim is true.
A person can reasonably question whether an institution has been honest about a particular event without concluding that every opposing account must be part of a coordinated deception.
Likewise, someone can find a conspiracy video fascinating without accepting the creator's conclusions.
The distinction is between curiosity and conviction.
Curiosity asks, What evidence supports this?
Conviction without adequate evidence says, I already know this is true. Now show me more reasons to believe it.
One invites investigation.
The other risks turning investigation into confirmation.
How to Tell Whether You're Following Evidence or Following an Algorithm
Fortunately, the psychological research offers more than warnings.
Studies have found that even simple reminders to consider accuracy can improve people's ability to distinguish reliable information from misinformation before sharing it.
A 2022 analysis in Nature Communications examined 20 experiments involving 26,863 participants. Prompts directing attention toward accuracy reduced intentions to share false headlines.
That suggests a useful habit: interrupt the automatic reaction before accepting or sharing a claim.
When something online feels unquestionably true, try asking yourself:
1. Where did this claim originate? Is there an original document, recording, experiment, or verifiable observation? 2. Are these genuinely independent sources? Or are multiple creators repeating the same unverified story? 3. What evidence would change my mind? Can I identify a realistic observation that would show the claim is wrong? 4. Have I seriously considered the strongest contrary evidence? Not a caricature of the opposing position, but the best-supported explanation available. 5. Would I accept this standard of evidence if the conclusion were reversed? Or am I applying different standards depending on what I want to believe? 6. Am I responding to evidence or to how the story makes me feel? Excitement, suspicion, anger, and certainty are experiences, not measurements of truth.
None of these questions require abandoning curiosity.
They protect it.
The Most Dangerous Words Might Be "I Just Know"
Most of us have been wrong about something we once considered obvious.
We've misjudged people, misunderstood situations, trusted bad information, and confidently predicted outcomes that never happened.
Being wrong is part of being human.
What separates thoughtful skepticism from unquestioning belief isn't whether someone ever makes a mistake.
It's whether they're willing to recognize one.
The internet has given us extraordinary access to knowledge. But it has also given us an unprecedented ability to surround ourselves with information that feels reassuring, exciting, or validating.
And when recommendation systems repeatedly introduce us to content that supports what we already suspect, we can begin confusing the consistency of our feed with the consistency of the evidence.
The next time a video convinces you that you've uncovered a hidden truth, consider something.
You may genuinely feel that you've connected the dots.
You may be absolutely certain.
You may even wonder how everyone else could be so blind.
But remember that other moments in your life when you felt equally certain have sometimes ended very differently.
Your certainty is real. That doesn't mean your conclusion is right.
And an algorithm that keeps telling you what you want to hear may be the last place you'll discover the difference.
Sources and Further Reading
1. Ye, S., et al. (2026). Systematic review and meta-analysis of the evidence for an illusory truth effect and its determinants. Nature Communications. https://doi.org/10.1038/s41467-026-70041-x
2. Binnendyk, J., et al. (2022). Intuition, reason, and conspiracy beliefs. Current Opinion in Psychology. https://pubmed.ncbi.nlm.nih.gov/35785571/
3. Haroon, M., et al. (2023). Auditing YouTube's recommendation system for ideologically congenial, extreme, and problematic recommendations. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.2213020120
4. Biddlestone, M., et al. (2025). Reasons to Believe: A Systematic Review and Meta-Analytic Synthesis of the Motives Associated With Conspiracy Beliefs. Psychological Bulletin. https://doi.org/10.1037/bul0000463
5. McLoughlin, K., et al. (2024). Misinformation exploits motivations to share outrage online. Science. https://www.kellogg.northwestern.edu/academics-research/research/detail/2024/misinformation-exploits-motivations-to-share-outrage-online/
6. Pennycook, G., and Rand, D. G. (2022). Accuracy prompts are a replicable and generalizable approach for reducing the spread of misinformation. Nature Communications. https://doi.org/10.1038/s41467-022-30073-5