Как читать медиа?

How to check information and avoid becoming a victim of fake news

How to check information and avoid becoming a victim of fake news

Explained in 10 cards

TL;DR (explained in 2 paragraphs)

Verification of information is necessary for three reasons. Firstly, the amount of information consumed is growing, and it is important to understand which sources are the most reliable and authoritative. Secondly, an error can occur at any angle - from the source of information to its final consumer. Thirdly, 90% of messages are informational noise, and we need to isolate the signal.

It is impossible to establish the truth. This is especially true when it comes to current events and an attempt to understand the motives of stakeholders. But you can vaccinate yourself against false information by correctly asking questions and judging the incoming information.

How is information distributed in the modern world?

Short answer: fast, varied and uncontrollable. In addition, each link in the chain may fail.

More details. In the pre-digital era, the main distributor of information was mass media. States regulated the media system or built an effective self-regulatory scheme through the institution of reputation and professional standards.

Such a scheme had only two weak spots. Firstly, it works on a relatively small number of participants. While organizing your newspaper or television channel was difficult and expensive, it was easy to regulate communications. Secondly, the faster the pace of life and the more diverse the formats of the transmitted information, the more difficult it is to monitor the quality of the media.

In digital reality, each participant in communication can not only change the message, but also pass it on to an unlimited number of recipients. In this case, the transmission circuit may fail anywhere.

Simply put, the source, the journalist and even the consumer can make mistakes and noise in the message. The latter often does not fully understand the topic or does not take into account information that he does not like. In a digital environment where algorithmic tapes are tailored to your interests, doing it all is simpler and easier.

Why should the information be double-checked?

Short answer: So that the world in our head can match the real one.

More details. All news and information can be divided into two large groups: better to know (it would be nice to know) and need to know (you must know). Information from the first group entertains you, but has little effect on your life. For example, the real news “In Kostroma, the minibus driver left passengers for the sake of shawarma” is unlikely to change you, even if you are in the Kostroma minibus.

Another thing is if you do not know who to vote for in the election or work with the currency and want to know the dollar exchange rate in the past, present and future. In this case, the need to know information is more useful, the more relevant, more detailed and more accurate.

Unfortunately, the actors who are busy in the news performance are telling the truth and nothing but the truth. For example, the Canadian newspaper The Toronto Star analyzed in mid-2018 the veracity of 1.3 million words of Donald Trump uttered and written during his presidency. They found nearly 2,000 false allegations (or, more precisely, 68,928 false words). More than 100 million Americans are likely to take part in the 2020 new elections. Every lie of the current president can push them to make the wrong choice.

The more confident people are that the information in the news is true, the more justifiably media consumers will vote in the next election. In addition, personal finances often depend on the truthfulness of messages, as well as professional growth - it is bad when all you use are unreliable sources.

Unfortunately, there is so much information that it simply cannot be accurate and truthful. We need to develop a habit of media hygiene and clearly separate in our heads the funny news about reptilians that have captured the Earth and the less funny news about a possible tax increase, to separate fakes from the need to know section can cost you too much.

What are fakes and what is fake news in general?

Short answer. Surprisingly, there is no exact definition of fake news, the term means a mix of concepts.

More details. In everyday life, it is believed that fake news is necessarily a lie framed as news, but this is not entirely true, because often we cannot establish the truth even in the best examples of the news genre.

It is easiest to define from the contrary. If the news reports something important and useful, influences the society and decisions of its members, then fake news is messages with reduced value for society.

Obviously, reduced value or benefit does not necessarily mean social harm. It is unlikely that there will be a fan of healthy food who has never dreamed or did not take a break for a burger that expires in fat.

The same with news. Some fake messages can even be useful, others can hardly be called fakes - they have zero value for our future, like soda with a sugar substitute. Still others are, on the contrary, destructive.

Let’s narrow down some fakes and half-fakes.

Conditionally useful or harmful should be attributed:

  • News containing unintentional errors. For example, the source made a reservation, we accurately quoted his mistake, and the reader was misinformed as a result;

  • Satirical news. Sometimes you can draw attention to an important issue, bringing the situation to the point of absurdity. A satire is necessary for a healthy society, even if it masquerades as real news. In addition, satirical publications make other media more attentive to the selection of news (for example, a lot of fictional news by Panorama was reprinted by real media).

  • Trendy collections. Among the key values of journalism is a balanced and objective coverage of what is happening. But there are situations when, for example, it is more important to put forward the list of children who died from lack of medicines than to look for economic arguments against subsidizing purchases. A tendentious selection is a double-edged weapon that can be used for good, although it is much more often used for evil.

Neutral fakes:

  • Errors - technical, concept and others. Most often these are journalistic mistakes. Millions are confused with billions, surnames and positions are distorted, terms are incorrectly translated. Potentially, this can be harmful, but usually only brings confusion to the minds and quickly corrects itself.

  • Info-noise - news with zero information value. Often these are reports of cats returning home, spreading on Twitter a cute gif or little-known policy, who promised, in case of victory, to achieve world peace. Info-noise can inspire or evoke pleasant emotions, but it is often used by political strategists in order to distract readers from the really important news. The Info-noise column, for example, was launched by RTVI, thereby flagging such unnecessary news and contributing to the media literacy of viewers.

  • A little cheating or manipulation. Sometimes the manipulation of public opinion can play a neutral or the same socially positive role as satirical news. When the exalted dances of the hosts begin in Malysheva’s program, designed to hamper the audience with the importance of vaccinations, it is easy to close their eyes to many in consistencies, including the fact that everyone during the musical number is vaccinated with the same huge fake syringe.

Finally, consider the third, most unpleasant class of fakes. Actually, they are usually called fake news:

  • Direct lies. Sources quite often may not be wrong, but intentionally mislead. Journalists, in their own interests or following a distorted editorial policy to please the owner, may also violate the principles of balanced coverage.

  • Clickbait and commercial fakes. Useful information is either missing or hidden. Little-known experts boldly comment on the news sucked from the finger, and comments are put in the headline. There are many such fakes, they clog the reader’s head without bringing anything good. A large number of relevant headers can be gleaned, for example, from the song "Horse" of the group "Bricks". As for commercial fakes, they usually represent real news, developed and disfigured by marketers towards their own product. The fact that a can of soda, supposedly first reduced to 0.25 liters, has actually existed in this volume for several years in other countries will not turn your life upside down. Spending time checking such statements is foolish, but recognizing what is useful and what is not is not easy.

  • Propaganda. One of the most disgusting features of a commercial fake is the budget for promoting inaccurate information. Even worse is the same, but performed by the state, because businessmen spend their money, and the state - the money of taxpayers. Propaganda news is usually one-sided, sowing discord, and absolutely certainly are not considered carefully. For this reason, in a number of countries, the internal broadcasting of state radio stations and television channels is prohibited or severely limited. 

Is there a way to determine if a fake is in front of you or not?

Short answer. No, but there are useful methods.

More details. The easiest approach is to carefully read the text and find out whether each of the facts really took place. This is very time-consuming, requires high qualifications and is not suitable for the average user. Sometimes the computer forces the text to be analyzed, but this often only leads to the blocking of "yellow" news and false positives due to the list of what can be called stop words.Поэтому специалисты применяют иные подходы.

Firstly, scientists found that fakes spread differently than regular news, but special skills are needed to notice this. However, if you see that a particular news item is spreading too well, it’s worth a first warning to make this news. This and similar approaches are used in social networks to slow down the spread of fakes.

Secondly, you can refuse to determine whether it is a fake or not, and simply restrict the distribution of any messages, as WhatsApp did. The most popular messenger on Earth has banned forwarding messages first to more than 20 recipients, and then to five altogether. However, the likelihood that a particular news has more than five recipients is small. Considering that messages can be sent to groups, which contain no more than 256 people, the audience of one message is limited to approximately 1300 users.

Thirdly, others say, inaccuracies can be missed in any message. It’s much more important for us to increase the likelihood that you are reading quality articles. Then the attack area available to the fakes will be significantly reduced. The evangelist of this approach is Frederick Phillou, who developed the aggregator deepnews.ai based on the determinant of quality texts.

Deepnews.ai uses a scoring model similar to the one your bank uses to determine whether to give you a loan and under what conditions. It’s not very important for the bank whether you get tired at work, it’s much more interesting how much you get. Phillou's scoring model (first simple, then upgraded based on machine learning algorithms) does not take into account the content of the article. She does not verify the facts. Instead, she reasonably believes that the creators of fakes on average spend less energy and resources on their "journalism" than the real authors.

And, is it working?

The Phillou method works in 80% of cases, but using the principles laid down, a person can bring this indicator to almost 100%. Here are five simple principles by which you can determine how high quality your article is.

ATTENTION. Poor design quality does not automatically mean that the note is false. And vice versa, quite well-written notes can be rubbish. The following are creatively revised principles - not all of them can be programmed, but each is accessible to humans.

Principle one: check the platform. Make sure you know the site or account on which the message is posted. Take a closer look if this is a real site, or just similar to it. Well, if the note has an author, and you know his reputation. It’s even better if the section “About the project” says who finances the site - so you can understand if the editorial staff has a conflict of interest when covering the event. For example, state-run media are unlikely to criticize the current government, and the television channels belonging to any businessman will cover the activities of this businessman independently and dig for him if he gives cause.

Principle two: check the sources. It’s good if the note has two or three independent sources. This greatly reduces the likelihood of a mistake or rumors of news. If the sources are well-known agencies, and the site you met for the first time, try to find the original message. If a note refers to a post on a social network and at the same time quotes pieces from there, it is almost always better to read the post itself and form an opinion - otherwise, quotations may be distorted.

The third principle: allow only real experts into your head. Make sure that commentators represent different sides of the conflict, if any.In addition, share your feelings for the source and its competence. The singer may emotionally advocate for or against the death penalty, but such a statement does not carry information content.Commentators should have relevant experience. If the senator comments on the scientific part of the news on nanotechnology, this is strange, because usually senators are better versed in law and public administration. A complete analogue of the situation is that if the working group on amending the Constitution consists of television signs that did not read the document and comments on it, something goes wrong.

Fourth principle: the more work that’s been put into something, the better. Fake creators rarely care about quality, their works are short-lived. Therefore, the amount of labor invested is the best indicator of non-fake. The more facts, good sources and commentators, illustrations, diagrams, screenshots and videos in a note, the better. Very rarely, well-designed multimedia stories turn out to be fakes, and when this is the case, the distortions are immediately noticeable: an unbalanced, evaluating tone of presentation appears, the author does not give two opinions and generally tries to push you to conclusions.

If a note answers all questions, does not try to convince you of something, but submits information remotely and immediately from several points of view, it is most likely true.

Fifth principle: seek and spread valued information. After reading the first lines and understanding the balance of power, look for context, useful additional information and facts that contradict your picture of the world. You will see an informational occasion in all the media, but useful texture is very rare.

  • Constantly ask yourself if you really need this or that news for life or professional development;

  • Never judge the headline or eyeliner of all the news; 

  • Never distribute the news without going to it and reading the entire article. If possible, show your attitude to the news during the repost and confirm it with facts. Then your friends will not have to go all your long way.

Is it really necessary to suffer like this every time? Isn't there someone who needs to verify the information?

By 2020, the world is arranged in such a way that literally all process participants are involved in checking information:

  • Faced with pulling quotes out of context and distortions, newsmakers are asked to send their words for approval or give comments only in writing.

  • Journalists take special courses to identify fakes;

  • In different countries of the world, for years, media consumers at schools have been conducting media literacy classes for years along with more familiar subjects.

Social media and techno-corporations also release special tools to combat fakes. We have already cited the example of WhatsApp, which restricts the distribution of news, because fakes are characterized by fast distribution. But, for example, on Twitter, fakes spread even faster than regular news.

Jigsaw’s Alphabet recently introduced Assembler, a tool that checks images for seven of the most common image manipulation techniques, including diphakes, a technology that allows you to stretch anyone’s face into any video.

In China, WeChat and Weibo messengers are actively fighting coronavirus fakes. Technological platforms now have special departments of fact checkers checking messages.

Weibo regularly publishes and debunks myths, and also distributes blog checks from 11 millionth Wuhan to cut off impostors.

In WeChat, where there is a platform of gadgets, one of them checks rumors about a coronavirus and marks them as true, doubtful or false. By February 1, this application was launched 350 million times.

I see a message on social media. How do I make sure it’s not a fake?

Short answer. No easy way. But you can significantly reduce the likelihood that you see a fake.

More details. Here are some signs of a non-fake. Remember that none of them guarantees the truth of the information provided, but the more they are in the message, the more likely it is that you have good news.

Account verified. Most social networks (Facebook, Twitter, Telegram) have a verification program, better known as a blue checkmark. People with a blue checkmark do not necessarily tell the truth, but they are usually famous for something, and their statements are supported by their reputation.

Message - not a screenshot. Very often sites publish screenshots of certain posts from social networks in order to avoid the situation when the post is deleted, and part of the article will be based on an inaccessible message. This complicates the check - because if the real mechanism of embedding or at least repost is involved, we can go to the original message with one click.

  • The message does not contain errors or typos, especially in the names of companies and actors. This kind of distortion is a frequent sign of satirical publication.
  • You know the author’s reputation. The author has not previously published unverified information.
  • The author is a specialist in the topic that he raises in the message. If an actor reposts a message about a viral threat from China or the opening of a vaccine, there is a chance that he does not understand what he is writing about.
  • You see other messages on the same topic from trusted sources, they confirm this and do not refer to it.
  • The tone of the original message corresponds to the tone of the author. In other words, he did not reprint someone else's text, but wrote his own, having invested his strength.
  • There is no commercial component in the message.

Can I get an example?


In February 2018, one of the Twitter users “reminded” the others that before his presidency, Donald Trump allegedly wrote that if the Dow Jones fell more than 1000 points in a day, the current president should be pushed into the cannon and shot in the sun.

The joke got out of hand, despite the fact that it is obviously a fake:

  •  The retweet was not used, but a screenshot.

  •  If you try to flip Trump's twitter before February 25, 2015, there is no such tweet

  •  Instead of the Dow Jones, fake Trump writes Dow Joans.)

Why was the fake taken for the truth?

  • He was posted by a person with a verification tag (Shaun Usher - author of a series of Letters of Note books, collections of noteworthy letters; it was hard to expect hoax from him);

The screenshot shows a large number of retweets and likes, supposedly testifying in favor of the reality of the tweet.

Last question. Now I can identify fakes. Is this sleeping prime minister's photo real? Is this a cat video that I see as a gif?

Short answer. It is irrelevant.

More details. Factchekers are literally drowning in a huge number of fakes and mistifications. Fortunately, 90% of news releases do not require verification, because they do not affect real life.

In the era of fakes, post-truths and propaganda, one of the most effective methods of misinformation is to drown the target audience in information noise. Not so much to distract her attention from the important, how much to spend this attention in advance on minor news occasions, to dilute the agenda.

This very effective technique can be avoided in only one way:

  • Do not be interested and do not refer to those facts and news that you have not verified and which do not require your verification. Cats are cute, even if generated on a computer. The photo of the prime minister is unlikely to change your attitude towards the government - it is better to analyze his affairs or his property.
The author of the article: Alexander Amzin, an independent medical consultant, host of the TV channel @themedia, author of the books "Internet Journalism News" and "Internet Journalism". For a long time he taught at the Moscow State University's faculty, worked with dozens of Russian media, and was a product manager for The Bell.

Что еще почитать