How to Detect Misinformation and Fake News

0
12
Even seemingly legitimate sources can faked online.

These days untrue accounts of events, manipulated images, misinformation and intentionally deceptive news articles are spread like wildfire across social media platforms. These are usually shared by well-intentioned people unaware they’re sharing inaccurate information and are a growing concern.

As violence in and around Gaza continues to intensify, and the Ukraine war drags ever onward individuals are increasingly turning to platforms like Twitter (“X”) for the latest updates on the conflict. However, they are confronted with a deluge of false informationā€”outdated videos, fabricated images, erroneous reports, and outright lies.

The swift spread of readily available AI tools further complicates the situation, making it extremely difficult for even some researchers to immediately spot the fakes.

Determining the truth

In the face of such rampant misinformation, how does one determine what to trust, what not to share, and what to flag for tech companies?

Here is a basic guide to “Spotting fake news and information” for everyone to follow when consuming real-time news online.

UNDERSTAND THE ORIGINS OF MISINFORMATION

When reading news, always ask yourself, “who benefits from spreading misleading information” and acquaint yourself with the main narratives.

During elections, for instance, be vigilant for conflicting information, conspiracy theories, unsubstantiated allegations, and groundless concerns regarding voter fraud that may favour a specific political party or candidate. Propaganda is often wielded as a tool during armed conflicts, and it can emanate from official news releases or unofficial channels.

When something appears excessively sensational or appealing, irrespective of whether it aligns with your perspective, exercise caution and conduct further research. Misinformation is usually spread by individuals who wish it were true, rather than those who have checked to see if it is.

Those responsible may aim to incite their supporters, escalate tensions between opposing factions on an issue, or generally destabilise a populace. In fast-paced situations, where obtaining an accurate account of events may be challenging, wait until additional information becomes available before sharing or forming conclusions. Not all misinformation is malicious; some is crafted for amusement or to troll individuals, so maintain a degree of scepticism towards whimsical narratives as well.

ADOPT A MORE DELIBERATE APPROACH

Refrain from impulsively sharing content. Social media platforms are designed to facilitate the rapid spread of information, often leading users to share content before fully evaluating whether it is true. Irrespective of how compelling, enlightening, or anger-provoking a TikTok video, social media post, or YouTube clip might be, exercise restraint before sharing it with your network. Presume everything is questionable until its authenticity is verified.

SCRUTINISE THE SOURCE; QUESTION “VERIFIED” ACCOUNTS

Investigate who is disseminating the information. If it comes from friends or family, remain sceptical unless the source is a person present at the scene or a recognised expert.

If it originates from a stranger or an organisation, remember that a verified badge or widespread recognition does not necessarily make an account reliable.

Currently, numerous political commentators and prominent internet figures are sharing inaccurate information, necessitating a cautious approach towards every post. If the account sharing content is not the primary source, take the initiative to locate the original Facebook, YouTube, or social media account that initially shared it.

Failure to ascertain the content’s origin should raise a red flag. Be especially cautious of screenshots, which can be more challenging to trace, as well as content provoking an exceptionally emotional reaction. Disinformation exploits such responses in order to get your shares.

When scrutinising individual accounts, take note of the account creation date, usually listed in the profile. Exercise caution with very recent accounts, those created in the past few months, or those with only a limited number of followers.

To investigate a website’s legitimacy, utilise Google to discover the year of its creation. Search for the site’s name and click on the three vertical dots beside the URL in the search results to find the date of its first indexing by the search engine. Again, avoid very recent websites. Do not overlook the fundamentals: Perform a Google search for the person or organisation’s name.

RELY ON ESTABLISHED SOURCES

Conducting individual background checks on numerous social media accounts is time-consuming, particularly when content is being generated from multiple sources simultaneously. Instead, place trust in reputable news organisations, which are well-equipped to verify information and often report on the same videos or images captured by credible sources after confirming their origin.

Utilise dedicated news tools such as Apple News, Google News, or Yahoo News, which select reputable sources and offer built-in moderation. On social media platforms, create or locate lists of endorsed experts and outlets to follow, particularly in the context of the subject you are monitoring.

If you rely on X, formerly Twitter, for breaking news, be cautious when following established reporters from reputable outlets on the ground. Twitter’s modifications to its verification system have complicated this process and made it easy for scammers to clone accounts for real journalists. Do not automatically assume an account is trustworthy or validated by X merely because it bears a blue checkmark or was endorsed by the company’s owner, Elon Musk. Visit news websites directly or seek out trusted individuals on alternative platforms.

SEEK CONTEXT

Misinformation frequently surges before, during, and after major news events. Real-time updates often feature a stream of fragmented information from the field, including smartphone videos and personal accounts. Even if these posts appear authentic, they may still be bewildering or misleading, representing only a part of the broader picture.

Try to understand individual clips or narratives within the context around the events.

Incorporate insights from recognised experts in the relevant field, whether it pertains to foreign policy, cyber warfare, history, or politics. Additionally, consult online or television outlets that provide this contextual information.

IDENTIFY AI-GENERATED IMAGES

AI image generation tools make it more challenging to identify fabricated images, though not impossible. Consider these five indicators to recognise AI-generated images, focusing on details like hands, background elements, and inanimate objects that appear unusual.

Examine the hands: AI software occasionally generates human hands with an abnormal number of fingers or other irregularities, although the technology is improving rapidly.

Zoom in on inanimate objects within the image to detect anomalies:Ā Concentrate on elements like eyeglasses, fences, or bicycles, assessing them for distinctive flaws.

To determine if an image is AI-generated, scrutinise writing on objects such as street signs or billboards and verify whether it appears reversed or nonsensical.

Analyse the background: AI-generated images may exhibit blurred or distorted details, particularly in the background.

Assess whether the images appear excessively glossy or artistically stylised. Some AI-generated images of real individuals may have exaggerated, stylised features or depict individuals with plastic-like facial appearances. Employ these indicators alongside other signs of misinformation if you remain uncertain. Is the source reliable? Are there multiple reports corroborating the same event?

VET VIDEOS AND AUTHENTIC IMAGES

If you wish to conduct more in-depth assessments of unverified reports, commence with an extensive guide on how to scrutinise videos. Look for multiple edits, and abrupt cuts, or carefully listen to the audio. You may utilise third-party tools, such as InVid, to verify the authenticity of videos. This process may prove more challenging for live-streamed videos, as seen on platforms like Twitch or other live social media options