Conscious use of the media – especially in the era of artificial intelligence – protects against disinformation and hate. On the occasion of the World Media and Information Literacy Week announced by UNESCO, we remind you how to approach the content disseminated responsibly, especially on social media.
Most Poles indicate a number of threats arising from the Internet, including hate speech, hate speech, and disinformation. Nearly one third admit that when using the Internet they are afraid of disinformation and fake news. These are the conclusions of the August study on cybersecurity in Poland, from which: report published on October 3 by the Digital Poland Foundation. An earlier study from February this year showed that, according to 82 percent. Poles, the scale of dissemination of false information has increased over the past ten years; 84 percent respondents admitted that they had encountered fake news.
READ ON KONKRET24: Poles vs disinformation
As we observe at Konkret24, this phenomenon is getting worse. Year by year, it is becoming more and more difficult to eliminate fake news and broadly understood disinformation from the messages being spread. For example, 14 years after the Smolensk disaster, 29 percent of Poles surveyed in February still believed it was an attack. And this despite what has already been revealed by professional media evidencethat no teams of specialists and experts investigating the causes of the disaster have confirmed the theory of an attack.
From October 24 to 31, World Media and Information Literacy Week is held – an annual event organized by UNESCO. The aim is to raise awareness of the importance of a responsible approach to content spread in the media, especially social media. On this occasion, we would like to remind you what to do to avoid succumbing to disinformation and contributing to the spread of cyber hate.
The basic rule: do not uncritically approach everything that reaches us through traditional media, social networking sites or instant messengers.
Reliable information on the Internet: how to recognize it
Before you comment or share information you come across online, including on social media, check whether it is reliable and true. How to go about it? Start by identifying the source.
See which site published this content: Have you heard of it before? what other content does he publish? If the website does not contain data about its authors, if the texts are not signed with their name and surname, if there is no editorial footer – such a website is not very reliable.
Read the text carefully. Check whether it includes experts in a given specialization, e.g. scientists with achievements and publications on the topic on which they speak. If the comments are anonymous, the text lacks facts and specifics, references to sources, and the text is dominated by evaluative content – don't trust them. In order to weaken the vigilance of recipients, real information is often mixed with manipulation or fake news on such websites.
Distinguish opinions from facts. The opinion is subjective and contains evaluative vocabulary and assessments. Information about facts is objective, specific, its description is neutral and non-judgmental. If facts and opinions are mixed in the text, or opinions are presented to the reader as facts – do not trust such a message.
Also pay attention to the publication date. False narratives have the stronger impact the more they concern current events – what the recipient is currently living. Therefore, it is important to check when the material was published. Because it often happens that when, for example, a new armed conflict begins, old articles, recordings and photos are published as current ones.
Be wary of messages posted on anonymous social media accounts, even if they go viral. Be especially careful if such a message lacks a source or the source provided does not appear to be reliable.
Be aware that not everything you read and see on the internet, especially on social media, is true. So use information from reliable sources. These include professional media editorial offices, including: news portals (e.g. Tvn24.pl, Onet, Wirtualna Polska), news agencies (Reuters, Associated Press, Agence France-Presse), opinion-forming dailies and magazines, professional news broadcasters and televisions, and websites of such media. If a given piece of information has not been published by any large, opinion-forming media, approach it with caution.
Control your emotions on social media
Many of us comment or share content on social media under the influence of emotions, without considering whether it is true.
Creators of many disinformation messages build narratives on strong negative emotions, including fear. They want recipients to feel threatened and thus become more susceptible to what they read or see online. Fear also makes us automatically want to warn others, so we share alarming materials more often. All this contributes to the spread of, for example, war disinformation.
Therefore, before you react to a post or other material on social media, answer the following questions: What emotions does it evoke in me? Is what I share or comment on true information? It's not your goal to mislead others, is it? Your reaction to a potentially false post can increase its visibility to others – social media algorithms work in favor of content that triggers reactions from users.
If you are not sure whether a given information or entry is true, do not comment, do not share or react. If you suspect that the information is false, report it to fact checkers. At the end of the text we explain how you can do this.
Photos, video and audio in the age of AI. How to distinguish real from generated
Artificial intelligence is already present in everyday life. Many of us now unlock our phones with Face ID, rely on virtual assistants, and use AI-powered chats. Increasingly popular and advanced tools based on artificial intelligence influence the way people use information, especially available on the Internet.
You may have used information provided by an AI chat that turned out to be the result of an AI hallucination. Or maybe you were fooled by an AI-generated image because it looked like a real photo. Widely available generative AI creates texts, images, music, sounds and videos. It offers many opportunities, but also poses threats. AI is already used to create disinformation content, promote unreliable sources, and contributes to the multiplication of discriminatory prejudices.
READ ALSO IN KONKRET24: Trump and black voters. Presidential campaign with artificial intelligence in the background
So how can you distinguish a creation of generative AI from an authentic photo, video or audio recording? Here are some useful tips and tools.
Photo verification. Pay attention to details. Do you see the background elements melting, the characters have unnaturally smooth skin, their bodies are disproportionate? These are typical errors for graphics generated by AI-based tools. Pay attention to glasses, faces, eyes, hands, fingers, ears – mistakes are most likely to occur when AI generates them.
Publicly available online tools such as: Illuminarts, DeepFake-O-Meter, AI or NOT and Hive Moderation. Although the results provided by these tools are only a hint. It is difficult to judge based on them whether something is or is not created or manipulated using AI – but they are helpful.
Video verification. Watch the recording several times. Pay attention to details. Look for objects whose behavior defies the laws of physics, e.g. strands of hair that dissolve in the air or sea waves that, instead of hitting the shore, break in the other direction. Pay attention to body movements and the appearance of the people in the recording. Unnatural way of moving – e.g. minimal or no body movement; the body does not move when the head moves – these are typical AI errors. Another clue is the shiny, artificial appearance of the people's skin in the footage. Tools that can help you evaluate video include: TrueMedia.org, DeepFake-O-Mete, Deepware, Deepfake Total.
Audio verification. Listen carefully and analyze. Unnatural pauses and intonation, repetition of words – all this is suspicious. Check official sources – whether a given politician has also published the same voice recording there. The following will help you check audio recordings: AI or NOT, Hive Moderation and Deepfake Total.
Iran, China and Russia are trying to influence the US election. A flood of false information on the InternetMarcin Wrona/Fakty TVN
You are not alone with disinformation. Fact-checkers will help
Every day there is disinformation and fake news – including: those related to the war in Ukraine, the refugee crisis, climate change, election campaigns – are monitored by fact-checkers and specialized teams of journalists around the world, including in Poland. Verified messages and exposed manipulations can be found on the website Konkret24.
The results of fact-checkers' work are available on the Internet. They can be found, among others: In Google Fact Check Explorer – a database indexing only fact-checking articles from editorial offices from around the world. However, it is limited to fact-checks by editorial offices cooperating with Google. There is a similar international database The Database of Known Fakes (DBKF) created by WeVerify and developed thanks to EU funds. In these sources, you can check not only whether a given politician or public figure has previously been misleading and on what topic, but you can also analyze whether a given thesis or topic has already been verified by fact-checkers.
Verified messages and manipulations that are disseminated in the Polish media, instant messengers and social networking sites can be found, among others, on the website Konkret24.
If you are not confident in your ability to verify information, rely on professionals. You can report suspicious content – information, photos, videos – on the Konkret24 website. Just click a button “Report for review”. You can also report it via our social media – we are on Facebook, Instagram and X – or by e-mail at: Konkret24@tvn.pl.
Main photo source: Shutterstock