How to Spot Anti-Government Extremism in Someone's Posts

Most accounts that tip into anti-government extremism did not start there. Scroll back far enough and you find ordinary stuff - a job, a team, complaints about traffic and grocery prices. Then the complaints narrow. The government stops being something to argue with and turns into something to defeat. A meme praising an attack. A repost calling an official a traitor who deserves a rope. By the time it reads as a whole worldview, it has usually been building in public for months, one post at a time.

That slow build is useful if you are trying to read someone before you trust them, because it leaves a trail. You do not need private access or a hunch. You need to read what a person has already chosen to publish, and to read enough of it to tell a bad afternoon from a settled pattern. This guide walks through the signs, where they show up on the platforms people actually use, and the line that separates loud dissent from something worth taking seriously.

It rarely looks like a manifesto

People expect extremism to announce itself. It mostly does the opposite. The dangerous stuff hides in tone and repetition: a joke that keeps landing on the same target, a "just asking questions" post about who really runs things, a steady diet of reposts from accounts that treat elected officials as enemy combatants. Any one of those, on its own, is noise. The signal is when they stack up into a theme the person returns to, week after week, in front of everyone they know.

None of this is fringe anymore. About 56% of Americans say they have experienced online hate or harassment in their lifetime (ADL, 2024), which is the kind of climate where the harshest talk gets normalized fast and a reader learns to shrug it off. Reading closely is how you stop shrugging.

How to spot anti-government extremism, post by post

A few things recur often enough to be worth naming. None is a verdict by itself; together, and repeated, they change the picture.

• Glorifying political violence. Praising an attacker, framing a riot as a heroic act, or wishing harm on named officials. Not "I hate this policy" but "someone should make them pay."
• Insurrection and civil-war talk. Treating armed conflict with the state as inevitable or desirable - "it's coming", stockpiling as a personality, a countdown energy to the whole feed.
• Sovereign-citizen pseudo-law. The claim that taxes, courts, licenses or police have no authority over them, dressed up in fake-legal language. It sounds eccentric until it is the whole account.
• Dehumanizing whole institutions. Every agency is a criminal syndicate, every official a puppet, every disagreement proof of a plot. When there is no legitimate opponent, only enemies, that is the tell.
• Amplification over authorship. Watch what a person reposts and praises, not only what they write. Quietly boosting extremist accounts all day is a position, even with no original posts of their own.

Where dissent ends and extremism starts

Here is the part that keeps you honest. Criticizing a government is not extremism. It is the normal business of a free country, and most people who post angrily about a law, a leader, or an agency never come close to endorsing harm. If you treat every sharp opinion as a warning sign, you will flag half the internet and learn nothing.

The line is not volume, it is violence. Dissent argues, protests, votes, and mocks. Extremism celebrates attacks, threatens specific people, or insists the rules of a shared society simply do not apply. Someone can loathe a president and still be a perfectly safe roommate, client, or date. What changes the read is a repeated move from "this is wrong" to "this justifies hurting people." When you see that turn show up again and again, you are no longer reading a critic. You are reading a pattern.

A checklist for reading an account

1. Open their public accounts on X, Facebook, LinkedIn, TikTok and Instagram - the posts anyone can read without following them.
2. Read a few months of posts in order, so you judge the pattern instead of one line pulled out of context.
3. Note what they repost, quote and praise, not just what they write themselves - amplification is a stance.
4. Watch for repeated glorification of political violence, insurrection or civil-war talk, and sovereign-citizen pseudo-law.
5. Separate ordinary political anger from a steady theme that endorses or calls for violence against real people.
6. If reading hundreds of posts by hand is too much, run a scan that flags extremist, hateful and conspiracy content and shows you the actual post to judge.

The honest limits

Be straight about what this can and cannot do. It reads public posts only - a locked account, a burner they keep off to the side, or a private group chat is out of reach, and a person who barely posts leaves little to read. It works best on active accounts, where months of public activity actually add up to a pattern.

And it is AI flagging content with the receipts attached, not a judge. Context is exactly where machines slip: a historian quoting a manifesto, a comedian doing a bit, reclaimed language, flat sarcasm - any of these can get marked when nothing was meant. That is the whole reason a scan puts the actual post in front of you instead of handing down a score. A clean result means nothing in the public posts stood out, not that a person is safe or vetted. This is a personal read of what someone already made public, not a background check or a consumer report, and it is not a factor in any hiring, tenancy, or credit decision - use a licensed provider for those. Read for yourself, and let the posts, not the label, make the call.