First of all, I took you to use "disagree" in the sense of politically disagreeing. You can identify alleged information as being false based on other, more rel

First of all, I took you to use "disagree" in the sense of politically disagreeing. You can identify alleged information as being false based on other, more reliable evidence independently of whether you agree or disagree with any political position it is used to support. Second, often the original source can be identified, and in certain cases it's clear when an original source is compromised or unreliable. For instance, sometimes the original source are mere bloggers or activists with no privileged access to information. Other times the source is a known misinformation outlet. For example, wareonfakes.com is registered in Moscow, has no identifiable sources or posters or anyone else taking responsibility, and is obviously not upholding any journalistic standards. Third, it is often forgotten that misinformation can also involve correct information - this is even the standard. The misinformation part is in skewing someone's perception of reality. This can be recognized with a bit of common sense. For example, reports of small, anecdotal incidents are irrelevant for judging anything. The "the lady who defied a soldier from country X" were spread both by Ukraine and Russia, and are irrelevant to evaluating the overall situation. Other reports are not credible from the start. A typical example was the "Ghost of Kiev" story, which doesn't even need fact-checking to tell it's main purpose was propaganda and the informational value was almost zero. Fourth, not all of this is hard to recognize, sometimes misinformation is spam in the traditional sense. For example, there used to be (or still are) a massive number of posts under Youtube comments to advocate the not-yet-closed Youtube channel of a vlogger of US origin who claims to be a journalist (for which the person has no credentials) and is embedded into the Russian military campaign. These kind of repetitive referrals to other outlets are spam under any definition of "spam", and violate Youtube's ToS. Fifth, you can recognize intentions by analyzing the texts and behavior of the actors, and analyzing their amplification networks (who repeats which Tweet, for example). Humans are fairly good at recognizing bad intentions, although it's arguably harder when you only have written text available. A social network analysis helps, since bad faith actors elicit different publication patterns than ordinary users. Last but not least, in the case of Russia's current aggressive war against Ukraine, it's not just about misinformation. Not everyone is aware of that, but in many jurisdictions supporting Putin's war can constitute a crime punishable under penal law, just like using the word "war" can constitute a crime in Russia. In my opinion, all posts in favor of Russia's aggression should at least be removed from social media. This shouldn't even be controversial, 141 countries have condemned this aggression which violates all international laws. It was not common during WW2 for US companies to give the German Nazi party free airtime on all broadcast channels. People could have their personal, pro-Nazi opinions in allied countries, but expressing them publicly had - and should have had - negative consequences. There is no reason to think this should be otherwise in the current conflict, which is essentially a large proxy war between NATO countries + Ukraine as defenders against the sole aggressor Russia. Not just misinformation should be stopped in this particular case, but also open expressions of support for Russia, displaying the swastika-like "Z" sign, and so forth. And, of course, the NSA and other intelligence agencies should hack back to destroy Russian propaganda outlets and take them off the net. However, I admit that this last point kind of deviates from the original discussion. In a nutshell, there are many indicators for spam and many indicators for misinformation, and you judge these in a way similar to how a judge would evaluate cases that primarily involve circumstantial evidence. Humans can do this reasonably well, algorithms are still bad at it.