Social-media stories can move faster than verified facts. Find the original source, check the evidence and verify the story before sharing it.

ISA Intel Briefing

Fact Checking Political Information

Fact Checking: Why Voters Today Have to Verify What They Read

By Matthew C. Parker | CEO & Director of Training, Independent Security Advisors LLC | EPTraining.us | Defending Democracy Initiative

Date Posted: September 24, 2026

Abstract

This briefing accompanies AI and Deepfakes: How Fake Videos, Voices, and Photos Can Fool Voters. That article looks at how artificial intelligence can create convincing fake media. This one deals with the next question: how can an ordinary voter, journalist, investigator, or protective intelligence professional check what they are seeing?

Social media has given us more information than at any other time in history, but more information does not automatically mean better information. A photograph can be real and carry a false caption. A ten-second video can leave out what happened immediately before or after it. A political claim can appear to come from thousands of independent people when much of the activity is being pushed by a much smaller group of accounts.

The answer is not to believe nothing. The answer is to verify what we are seeing before we decide what it means or help spread it. You do not have to become an intelligence analyst to do it. Most of the time, a few simple checks will tell you a great deal.

Bottom Line Up Front

When something important happens, social media often begins telling the story before all the facts are known. Pictures are posted, clips are shortened, political groups and commentators begin weighing in, and people who were nowhere near the event start repeating claims as fact.

The first question should not be whether the post agrees with our politics. The first question should be much simpler: Where did this information come from?

Find the original source when possible. Look for the complete video instead of the shortened clip. Check when the photograph first appeared. See whether another independent source confirms the story. When the claim involves an election, school, court, police department, emergency, or government agency, go directly to the responsible organization and see what it actually says.

For deeper research, there are tools that can help us see how information spreads, whether groups of accounts appear to be moving together, where photographs have appeared before, and when a website came into existence. Those tools can give us better information, but they still require judgment.

A tool can help us find evidence. It cannot do our thinking for us.

Your Social-Media Feed Is Not Reality

A social-media feed gives us information, but it is not a neutral picture of what is happening in the world.

Some posts become popular because large numbers of people are genuinely interested in them. Others get pushed by political groups, advocacy organizations, activists, commentators, influencers, news outlets, businesses, coordinated accounts, or automated systems. In many cases, several of those things are happening at the same time.

That creates a problem during a fast-moving event. One group may take a local incident and immediately connect it to a larger national political cause. Another group may take a short video from the same incident and use it to portray the people involved as dangerous, dishonest, radical, corrupt, or part of some larger movement.

Large numbers of shares do not make either story true. They tell us that the story is moving. That is an important distinction.

We should also be careful about assuming that online activity is either completely spontaneous or completely manufactured. A real event can produce genuine public anger while organizations are also helping amplify it. A political group can promote a demonstration without every person attending being controlled by that group. A story can begin organically and later be picked up by national organizations, influencers, or coordinated networks.

Reality is often more complicated than the version that fits neatly into a social-media post.

Start With the Original Source

Before using any special software, start with the most basic form of fact checking. Find out where the claim began.

If you are looking at a screenshot, find the original post. If you are watching a short video, see whether a longer recording exists. If a politician supposedly made an outrageous statement, find the full speech, interview, hearing, or press conference. If a photograph supposedly shows something happening today, find out whether the picture existed before today. This is basic source work, and it solves a surprising number of problems.

A screenshot can leave out the account name, date, previous comment, or correction. A short video can begin five seconds after the part that explains what is happening. A quotation can be accurate word for word and still be misleading because the sentence immediately before or after it changes the meaning. The closer we can get to the original material, the better chance we have of understanding what actually happened.

Look for Independent Confirmation

The next step is to find another source that established the same basic facts independently. Ten social-media accounts repeating the same post are not ten sources. They may all be copying one person. This matters when stories move quickly. A claim can appear to have broad confirmation when hundreds of accounts are simply repeating the same original statement.

The number of times something is repeated tells us how widely it traveled. It does not tell us whether anyone independently verified it.

When the subject involves government action, public safety, schools, courts, elections, or emergencies, there is often an official source available. That source should be checked along with credible independent reporting. Official sources can be wrong, incomplete, or updated as new information becomes available. That is why the answer is not to depend on only one source. The goal is to compare what different sources can actually establish.

Seeing How Information Spreads

Sometimes we need to go beyond a single post and understand the larger online conversation. This is where network-analysis tools become useful.

Researchers, journalists, investigators, and protective intelligence analysts can look at connections among accounts, reposts, replies, hashtags, keywords, and other activity. Instead of seeing thousands of individual posts one at a time, the analyst can begin seeing the shape of the network behind them.

Indiana University’s Observatory on Social Media, or OSoMe, develops tools specifically for studying how information spreads and how coordinated activity can appear online. Its current tool set includes OSoMeNet and Coordiscope. (Observatory on Social Media)

These tools are useful because a social-media feed can make a conversation look like one enormous crowd. Network analysis may show something very different. It may reveal several separate communities, a few accounts acting as bridges between them, or a relatively small number of accounts responsible for a large share of the activity.

That gives us a better place to start asking questions.

OSoMeNet: Follow the Story

OSoMeNet is an Indiana University tool designed to visualize how information is being shared and spread across social-media networks. It can build diffusion and co-occurrence networks and show activity over time. The current platform supports sources including Bluesky, Mastodon, TikTok, and OSoMe’s historical social-media archive. (Observatory on Social Media)

In simple terms, OSoMeNet lets us step back from one post and look at the larger pattern.

Imagine that a political story appears to be coming from thousands of different people. A network may show that much of the activity began with a smaller number of accounts and then spread outward. That does not automatically make the story false. It tells us something important about how the information traveled.

For an analyst, that can lead to better questions. Which accounts were early in the network? Which accounts produced the largest amount of amplification? Did separate communities begin talking about the subject independently, or did one group feed information into another?

That is useful intelligence because it helps us understand the information environment instead of simply counting posts.

Coordiscope: Looking at Coordinated Activity

Coordiscope looks at another part of the problem. Indiana University’s Observatory on Social Media describes it as a tool for detecting potential coordination between social-media accounts. It compares patterns of account behavior and visualizes accounts that behave in similar ways. (Observatory on Social Media)

That can help when many accounts suddenly begin moving in similar ways.

Coordination itself does not tell us why the activity is happening. Political campaigns coordinate messages. Advocacy organizations organize supporters. Government agencies issue coordinated public information. Businesses conduct campaigns. Activists organize around causes.

The value of Coordiscope is that it can help identify a pattern worth examining more closely.

For a protective intelligence analyst, that can help distinguish between broad public attention and an online conversation in which a smaller network is doing much of the work. From there, the analyst can look at the accounts, timing, content, relationships, and other available evidence.

The tool points us toward the questions. The evidence still has to answer them.

Gephi: Building the Larger Network

Gephi is an open-source network-analysis and visualization program. It can handle connected datasets, identify clusters and central accounts, and turn large amounts of relationship data into a network map that is easier to understand. (Gephi)

For social-media work, an analyst might represent an account as a point in the network and use connections between those points to represent reposts, replies, mentions, or other relationships.

Once that information is mapped, patterns can become much easier to see.

One account may sit in the middle of a large network. Several separate groups may have only one or two accounts connecting them. A small group may produce far more activity than its size would suggest.

None of that tells us automatically what the people behind the accounts believe or why they are doing it. What it does is give the analyst a clearer picture of the information network and a better idea of where additional research should begin.

For intelligence work, that can be far more useful than simply saying that a topic had ten thousand posts.

Most Voters Need Much Simpler Tools

Most people do not need network-analysis software. If you are trying to decide whether something you saw on Facebook, X, TikTok, Instagram, or a political website is true, simpler tools will usually get you much farther.

The basic process is straightforward: check the source, search the claim, examine the photograph or video, look at the website, and see whether independent evidence supports the story.

The question remains the same throughout the process: Does the evidence match the story being told?

Fact Check Explorer and Fact-Checking Sites

Google Fact Check Explorer allows users to search published fact checks by topic, person, and image. Google explains that it provides access to fact checks published by outside organizations and does not itself create or endorse those fact checks. (Google Toolbox)

That makes it useful as a search tool.

If a political claim has already circulated widely, somebody may have already investigated it. Searching the main words from the claim can lead to earlier reporting, source material, and explanations of where the story began.

FactCheck.org, a project of the Annenberg Public Policy Center at the University of Pennsylvania, describes its mission as reducing deception and confusion in U.S. politics. Its published process emphasizes primary sources and applying the same accuracy standards to claims made by different political sides. (FactCheck.org)

Other established fact-checking organizations can also be useful, especially for viral rumors, recycled photographs, and claims that have already moved widely across social media.

The important part is to read beyond the rating or headline. Look at the evidence. Follow the source documents. See what was actually established.

Fact checking should help us understand the evidence, not simply give us another opinion to repeat.

Check the Photograph

A photograph can be completely real and still be used to tell a false story.

An image from another city, another country, or another year can be reposted with a new caption and presented as evidence of something happening today. No sophisticated artificial intelligence is required. Sometimes all it takes is an old photograph and a new sentence underneath it.

TinEye is a reverse-image search engine that can locate matching and modified copies of an image across its index. Its tools can help identify where an image has appeared, how it has been used, and whether altered versions exist. (TinEye)

Google Lens can also search using an image and return visually similar images and related pages from around the web. (Google Lens)

If a dramatic photograph appears in your feed, checking it with more than one image-search tool can sometimes reveal an earlier appearance almost immediately.

That may not answer every question, but it gives you something much more useful than a caption written by an unknown account.

Check the Video

Video should be treated the same way.

Find the original if possible. Look for a longer version. Check when it first appeared. Pay attention to whether the clip begins or ends at a convenient moment.

The InVID-WeVerify Verification Plugin provides a set of tools for checking online video and images. Among its functions are breaking video into keyframes, reverse-image searching those frames, examining contextual information, and using other verification tools in one place. (GitHub)

Breaking a video into individual frames is particularly useful because one frame may be easier to trace than the entire video. You may discover that the supposedly new video appeared online years earlier, was recorded in another country, or originally had a very different description.

The point is not to turn every voter into a digital-forensics specialist. The point is to take one additional step before accepting what somebody tells you the video proves.

Check the Website

Sometimes the picture and video are fine. The problem is the website publishing the story.

A website can give itself a professional name, copy the appearance of a news organization, and begin publishing almost immediately. That is why checking the history of the domain can be useful.

ICANN Lookup provides registration information that is publicly available for domain names. Since January 2025, ICANN has used the Registration Data Access Protocol, or RDAP, as the definitive system for delivering registration information for generic top-level domains. (Registration Data Request Service)

Registration information may show when a domain was created, which registrar handles it, and other technical details.

If a website describes itself as a long-established local news organization but the domain was created very recently, that is information worth considering.

It does not automatically mean the website is dishonest. It means we have another fact to compare with the story the website tells about itself.

That is what good verification looks like. We keep adding facts until we have enough information to make a reasonable judgment.

Do Not Let a Graph Replace the Investigation

This point matters especially in protective intelligence. Technology can make weak analysis look impressive. A network map with thousands of points and lines can fill a briefing screen and create the appearance that we know exactly what is happening. We may not.

If ten thousand posts appear around a political event, we still need to determine what those posts represent. They may reflect thousands of independent people. They may be heavily amplified by a smaller number of active accounts. Political organizations may be involved. Influencers may be driving activity. Automated systems may be present. The event may simply be attracting enormous public attention.

More than one of those explanations may be true at the same time. The network map helps us see the activity. Good analysis requires us to keep going.

For protection professionals, that matters because online activity can influence decisions about staffing, event security, crowd expectations, protective intelligence, and threats around a principal. Those decisions should be based on evidence rather than the size of a social-media number or the appearance of a complicated graph.

The same principle applies to voters. Find the evidence first. Then decide what the evidence means.

Why This Matters to the Defending Democracy Initiative

Media literacy belongs in ISA’s Defending Democracy Initiative because false and misleading information does not always stay on a computer screen.

The Initiative is concerned with threats, intimidation, political violence, protective intelligence, and the ability of people to participate in democratic government without fear or violence deciding who can take part. That includes elected officials, judges, prosecutors, candidates, election personnel, public employees, law-enforcement personnel, and citizens involved in the democratic process.

The mission is not to protect public officials from criticism. Government officials should be questioned, challenged, investigated, protested, and voted against when citizens believe that is appropriate. Those things are part of democratic government.

The problem begins when false information helps create real-world intimidation or danger.

A misleading photograph, false accusation, altered video, or manufactured story can place a mayor, judge, prosecutor, candidate, police chief, election worker, or other public official in the center of public anger very quickly.

Most people who become angry online will never threaten anyone, and political criticism by itself is not a threat indicator. Protection professionals still have to focus on behavior.

The environment changes when attention begins producing repeated unwanted contact, direct threats, publication of home addresses, research into schedules and travel, surveillance, attempts to defeat security measures, weapons-related behavior, or growing fixation on one identifiable person.

That is where media literacy and protective intelligence meet. The better we become at establishing what actually happened before spreading a story, the harder it becomes for false information to move real people toward real-world consequences.

The Bottom Line

You do not have to become an intelligence analyst to become harder to fool.

Start with the source. Find the original photograph or video when you can. Check the date. Search the main claim. Look for another source that independently confirms it. When the subject involves voting, schools, courts, police, public safety, government action, or an emergency, check the responsible organization.

Reverse-image tools can help identify old photographs. Video-verification tools can help trace footage. Domain records can tell you something about the website publishing a story. Network-analysis tools can help researchers see how information is spreading and which accounts are helping it move.

None of those things replaces judgment.

The internet allows all of us to see, react to, and publish information faster than at any point in history. Good judgment still depends on something much older: finding the source, checking the evidence, and being willing to wait when the facts are not there yet.

That is what being a fact checker really means. Verify what happened first, then decide what you think about it.

About the Author

Matthew C. Parker is the CEO and Director of Training for Independent Security Advisors LLC and EPTraining.us. A retired U.S. Army professional with more than 20 years of service, Parker’s career has included executive and dignitary protection, physical security, anti-terrorism, investigations, security training, and security program management in the United States and overseas.

Through ISA, Parker develops and teaches executive-protection and security programs for law-enforcement officers, military personnel, private-security professionals, and others responsible for protecting executives, elected officials, judges, prosecutors, and other at-risk individuals.

Parker is also the founder of ISA’s Defending Democracy Initiative, which examines the relationship between threats, intimidation, political violence, protective intelligence, public-official security, misinformation, and the ability of citizens and public servants to participate safely in democratic government.

Sources and Verification Tools

Indiana University Observatory on Social Media — OSoMeNet and Coordiscope. OSoMe Tools

Gephi — open-source network analysis and visualization. Gephi

Google Fact Check Explorer — search published fact checks by topic, person, or image. Google Fact Check Explorer

FactCheck.org — political fact checking from the Annenberg Public Policy Center. FactCheck.org

TinEye — reverse-image search. TinEye

Google Lens — image-based web search. Google Lens

InVID-WeVerify — video and image verification tools. InVID-WeVerify Verification Plugin

ICANN Lookup — domain-registration data. ICANN Lookup

#MediaLiteracy #FactChecking #OSINT #PoliticalMisinformation #ElectionSecurity #SocialMedia #ProtectiveIntelligence #DefendingDemocracy #EPTraining #IndependentSecurityAdvisors

Part 2. Fact Checking: Why Voters Today Have to Verify What They Read