What Is Media Literacy and How Do You Practise It on Any Online Claim

Stop before you share. That pause is the only habit that stops a scam or a falsehood before it reaches your contacts. Media literacy is the skill of judging whether a claim, image, video or piece of content is true, not whether it feels true. In Singapore, where messaging apps carry a relentless stream of forwarded-as-received material, the gap between feeling and knowing is where scams live. A working media literacy Singapore verification guide must be a process, not a set of principles. This is a five-question routine you run on any claim, image or piece of content before you believe it or send it on. The questions are simple. They are built on the same lateral reading technique professional fact-checkers use: open new tabs to check the source and the claims, rather than staying on the original page and letting its tone convince you. You are not aiming to become an expert in every platform. You are building a habit that takes under two minutes and works without special training. What follows is the repeatable, step-by-step verification process, framed for the Singapore context, where the regulatory landscape exists but is not the verification method itself.

Before you share anything, run it through five questions. They form a universal routine that works whether the content is a text from your aunt, a viral TikTok or a screenshot of a news headline. The five are source, evidence, context, intent and emotion. They are not equally weighted, and you will not always need all five. But the ones you skip are the ones that hurt you.

Question One: Source

Who is the original source? Not who forwarded it to you, and not the person whose name sits at the bottom. On WhatsApp and Telegram in Singapore, the forwarded-as-received label exists precisely because the platform cannot verify the chain. The person who sent it to you may be trustworthy. That does not make the original source trustworthy. Open a new tab and search for the source's name plus the word scam or the word false. Look for source transparency: does a fact-checker name the original source and show the evidence chain? A verdict without visible evidence is unverifiable by you. The source question also covers the officialdom halo: a piece of content that uses a government logo, official-sounding language or a URL that looks like .gov.sg but is not. The real Singapore Police Force and actual government agencies do use these channels. Impersonation scams are a top vector for fraud in this country. When in doubt, go to the agency's official website directly, not through the link in the forwarded content.

Question Two: Evidence

What is the evidence for the claim? Not the screenshots, not the dramatic video. The underlying evidence. For a news story, find the primary source: the court filing, the press release, the government gazette. For an image, run a reverse image search using Google Images, TinEye or Yandex. The tool may change. The principle is durable: check whether an image has appeared before in a different setting. A photo of a crowd in one city may be recycled to claim a protest in another. For a video, consider the possibility of a deepfake or a cheapfake. A deepfake is AI-generated synthetic media that replaces a person's face or voice; a cheapfake is a repurposed video edited to change its meaning. Both are on the rise because generative AI tools can now produce photorealistic synthetic media on demand. The evidence question is also about freshness. Recency bias makes us believe a claim because it references a current event, but the underlying information may be old or fabricated. Check the date stamp on the original source, not the date on the forwarded content.

Question Three: Context

What is the full context of the claim? A statistic without context is a lie. A quote without context is a misquote. The echo chamber is the enemy of context: algorithmic curation shows you content that reinforces your existing views, so you are more likely to see claims that confirm your bias and less likely to see the full picture. Confirmation bias is the cognitive side of the same coin. You seek out information that confirms what you already believe, and the platforms amplify it. To break the loop, use lateral reading to find the original study, the full press release or the verbatim quote. The context question also applies to the legal landscape. In Singapore, POFMA is the Protection from Online Falsehoods and Manipulation Act, a statute governing correction of online falsehoods. A POFMA direction is not the same as a fact-check, though both address falsehoods. The existence of the law changes the environment. It does not do your verification for you. The law is a backstop, not a checklist item. Similarly, the Personal Data Protection Act and the Personal Data Protection Commission govern how organisations handle your data. They do not tell you whether a claim is true. The context question is about the whole picture: the regulatory context, the platform context and the personal context of why this content was sent to you.

Question Four: Intent

Who benefits if you believe this? That is the intent question. A scam is designed to extract money or data, so the intent is clear. An impersonation scam poses as a government official, a bank officer or a family member to get you to act. The intent is greed. But intent is not always malicious. A well-meaning friend who forwards an urgent warning may have no intent to deceive. The effect is the same: the falsehood spreads. The verification bypass is the failure mode here. Sharing a claim with a comment like is this true? does not verify anything. It amplifies the claim to your entire network before anyone answers. The intent question forces you to stop and ask: if I share this, who benefits? If the answer is a scammer, you stop. If the answer is a political actor, you stop. If the answer is a platform that profits from your outrage, you stop. The intent question also covers the distinction between content moderation and legal action. Platforms remove content under their terms of service; the Singapore government acts under the law. The two are not the same, and confusing them is a failure mode. A platform takedown is not proof of a legal violation. A legal direction is not proof of a factual error. Both are actions with different standards and different appeal processes.

Question Five: Emotion

How does this content make you feel? Anger, fear, hope and disgust are the fuel of misinformation. The emotional override is a real effect: a claim that provokes a strong emotion short-circuits your verification step. You share it because it feels true, and the feeling is the point. The question is not whether the emotion is justified. The question is whether the emotion is doing the work that evidence should do. If you feel the urge to share immediately, that is your cue to stop and run the other four questions. The fact that a piece of content makes you feel something is not evidence that it is true. It is evidence that the content is designed to make you feel something. The five questions are not a guarantee. They are a habit. The more you practise them, the faster you get. Verification speed matters because the first few hours after a claim appears are the most dangerous. During that window, the claim spreads, the screenshots are taken, and the correction, if it comes at all, arrives too late to undo the damage.

When you encounter a claim that feels urgent, you need a concrete method. The following steps work with the five questions above, and they are built on the S.U.R.E. framework that the National Library Board promotes in Singapore. S.U.R.E. stands for Source, Understand, Research, Evaluate. It is a teachable verification sequence embedded in Singapore's school curriculum and public libraries, and it does not depend on any specific technology. The first step is to identify the source. The second is to understand the content. The third is to research the claim. The fourth is to evaluate the evidence.

Step One: Source Verification

Open a new tab and search for the name of the person or organisation making the claim. Add the word fake or scam to the search. If the claim involves a government agency, go directly to the agency's official website. For the Singapore Police Force, the National Crime Prevention Council runs an anti-scam website with a database of known scams and a helpline. ScamShield is the app and hotline that blocks scam calls and SMS and allows user reporting. It is a specific tool, not a general antivirus. It does not filter your entire internet. It does not replace critical thinking. It is a database of known scam numbers, and it is updated frequently. The update frequency matters because the scam numbers change daily. A number that was not in the database yesterday may be a scam today. The false positive rate exists, so a number that is not flagged is not proof of safety. The absence of a flag is not evidence of legitimacy.

Step Two: Evidence Verification

For image claims, use a reverse image search. Upload the image to Google Images, TinEye or Yandex. The tool may change. The principle is durable: check whether the image has appeared before in a different setting. A photo of a crowd in one city may be recycled to claim a protest in another. The reverse image search is not perfect. It fails on modified images. A simple crop or a filter can defeat it. A deepfake can defeat it entirely. For video claims, look for visual artifacts. Common AI-generated content indicators include asymmetric faces, unnatural hand positions and inconsistent lighting. But these indicators change with every model update. What is a reliable tell today may be fixed tomorrow. The only durable skill is to ask: could this be synthetic? If the answer is yes, treat it as unverified. The burden of proof is on the claim, not on you.

Step Three: Context Verification

Find the original setting of the claim. If the claim is a quote, find the full quote. If the claim is a statistic, find the original study. If the claim is a news story, find the original article. The setting is almost always more complex than the forwarded content suggests. A claim that appears to be about a single event may actually be about a pattern. A claim that appears to be about a current event may actually be about a past event that has been recycled. The context verification is also about the digital footprint of the claim itself. Search for the exact phrase of the content in quotes. If the same phrase appears on multiple sites with the same wording, it may be a coordinated campaign. If the phrase appears nowhere else, it may be a private note that has gone viral. The context includes the platform. Algorithmic curation on TikTok, WhatsApp, Telegram and Xiaohongshu differs from the better-documented Western platforms, but the principle is the same: the platform optimises for engagement, not accuracy. What you see is what you have engaged with, not what is true.

Step Four: Evaluation

Evaluate the totality of the evidence. This is where you make the judgment call. The verdict is not binary. A claim can be true, false, misleading, out of context or unverifiable. The evaluation step is where you decide which category applies. The user burden is real: each step costs time, and the total cost of verifying a single claim may be five minutes. That is five minutes you do not have if you are in a hurry. The cost of sharing a falsehood is higher. The cost is your credibility, your relationships and potentially your money or your data. The evaluation step is also where you consider the source transparency of the fact-checker. A fact-checker who names the original source and shows the evidence chain is more useful than one who simply issues a verdict. A fact-checker who publishes the reasoning is more useful than one who just gives a rating. The update frequency of a fact-checking source matters: a source that updates daily is more useful than one that updates monthly. Verification speed matters: a source that publishes a verdict within hours is more useful than one that takes weeks. Speed is useless without the evidence.

The S.U.R.E. framework is the National Library Board's contribution to media literacy in Singapore. It stands for Source, Understand, Research, Evaluate. It is a teachable sequence, not a technology. You can apply it to a printed newspaper, a broadcast segment or a forwarded piece of content. The framework is embedded in Singapore's school curriculum and public libraries, and it is the backbone of the Media Literacy Council's public education campaigns. The Media Literacy Council is a Singapore charity that runs public education campaigns on digital literacy. It is not a regulator. It does not enforce anything. It is not the same as the POFMA Office or the Personal Data Protection Commission. The distinction matters because a reader who confuses the two may expect the Council to act on a complaint, which it cannot do.

Source: The First S.U.R.E. Step

The source step in the S.U.R.E. framework asks you to identify who created the information. This is the same question as the source question in the five-question routine. The difference is that S.U.R.E. is a formal framework with a fixed sequence, while the five-question routine is a habit you can run in any order. The source step is where the lateral reading technique comes in. Lateral reading is the practice of evaluating a source by opening new browser tabs to search for information about the source or claim, rather than reading vertically within the source itself. The technique was studied by Wineburg and McGrew at the Stanford History Education Group, and their research showed that professional fact-checkers used it more effectively than self-described experts. The study is called Lateral Reading: Reading Less and Learning More When Evaluating Digital Information, published in 2016. The finding is that lateral reading is a skill, not a trait. It can be learned, and it should be practised.

Understand: The Second S.U.R.E. Step

The understand step asks you to make sense of the content. What is the claim, exactly? What is the tone? What is the purpose? Content designed to provoke panic is different from content designed to inform. Content that uses all-caps, exclamation marks and urgent language is a warning sign. Content that asks you to act immediately is a warning sign. Content that asks you to share it is a warning sign. The understand step is where you catch the emotional override. You pause and ask: why am I being asked to feel this? The answer is that the content is designed to make you feel this way. The understand step is also where you check for the verification bypass. If the content says is this true? then it is not a claim, it is a question. But the question is being shared, which means the claim is being amplified. Ask yourself: why am I forwarding a question I cannot answer?

Research: The Third S.U.R.E. Step

The research step is where the actual verification happens. You open new tabs. You search for the source. You search for the claim. You search for the image. You search for the video. You use the search engines, the reverse image search tools and the fact-checking sites. The research step is where you find the primary source. It is where you find the original study, the original press release, the original court filing. The research step is also where you find the corrections. A correction is not a failure. A correction is a normal part of the information ecosystem. The question is whether you see the correction. The correction blindness failure mode is when you see a fact-check or correction but dismiss it because it comes from a source you distrust. The research step is where you have to be honest with yourself: are you evaluating the evidence or the source? If the source is the Singapore Police Force, the POFMA Office or the Personal Data Protection Commission, the source is authoritative. Authority is not the same as infallibility. Even an authoritative source can make a mistake. The research step is where you check the source's evidence chain.

Evaluate: The Fourth S.U.R.E. Step

The evaluate step is the final judgment. You weigh the evidence. You decide whether the claim is true, false, misleading or unverifiable. The evaluate step is where you decide whether to share. The evaluate step is also where you decide whether to act. If the claim is a scam, you do not act. You report it to ScamShield or to the police. If the claim is a falsehood under POFMA, you do not share it. You wait for the correction. If the claim involves your personal data, you check the privacy policy. The evaluate step is where the Personal Data Protection Act comes in. The PDPA is the legislation that governs the collection, use and disclosure of personal data in Singapore. The Personal Data Protection Commission is the authority that administers it. The PDPA's core obligations are consent, purpose limitation, notification, access and correction, protection, retention limitation, transfer limitation and accountability. These obligations matter because they give you rights. You have the right to ask what data an organisation holds about you. You have the right to correct it. You have the right to withdraw your consent. The evaluate step is where you decide whether an app or a website is trustworthy with your data. The actual data handling is a question of trust, not just of law. An app that claims to protect your data may not, and the PDPC's enforcement decisions show organisations collecting, using or disclosing personal data beyond what is necessary or without adequate consent.

Images and videos are the most dangerous form of misinformation because they feel real. A photo of a disaster, a video of a politician, an image of a crowd: these are the raw materials of falsehood. The tools for creating them have become more accessible, and the results are more realistic. The first step in verifying an image is to run a reverse image search. The tool may change. The principle is durable: check whether an image has appeared before in a different setting. The reverse image search will show you where the image has been used before. If it has been used on a different story, the setting is likely wrong. If it has been used on multiple stories, it is probably a stock image or a recycled image. If it has been used on no other story, it may be new, which means you need to check the source.

Deepfakes vs. Cheapfakes

A deepfake is AI-generated synthetic media that replaces a person's face or voice. A cheapfake is a repurposed video edited to change its meaning. The distinction matters because the verification methods are different. A deepfake requires technical analysis. A cheapfake requires contextual analysis. A deepfake may be detected by looking for artifacts: asymmetric faces, unnatural hand positions, inconsistent lighting. These tells are not reliable, and they change with every model update. A cheapfake is detected by finding the original video. The reverse image search works on video frames. Take a screenshot of the video, then run the screenshot through a reverse image search. The results will show you whether the video has appeared before in a different setting. The key question is: does the setting of the video match the claim? If the claim says the video is from one city, but the original is from another, the claim is false.

AI-Generated Images

Generative AI tools like Midjourney, DALL-E and Sora can create images and video from text prompts. The results are photorealistic. The AI realism trap is real: you see an image that looks real, so you assume it is real. The tell is not always visible. The current generation of tools has improved, but there are still clues. Look at the hands. Look at the eyes. Look at the background. Look at the text. AI-generated images often have garbled text, even when the image is otherwise perfect. The clue may not be visible to you. If you cannot tell whether an image is AI-generated, treat it as unverified. The burden of proof is on the claim, not on you. The same applies to video. A video that is AI-generated may be indistinguishable from a real video. The question is not whether you can tell the difference. The question is whether you have verified the source enough to trust the claim.

Reverse Image Search Tools

Google Images, TinEye and Yandex are the three most common reverse image search tools. Each has a different database and a different algorithm. Google Images is the most comprehensive. TinEye is the oldest, and it is good for finding modified images. Yandex is the best for finding images that have been shared on Russian social media. The tool you use depends on what you are looking for. The principle is the same: you upload an image, and the tool finds where it has appeared before. The reverse image search is not perfect. It fails on modified images, cropped images and images that have been compressed. It is a good first step. The second step is to search for the setting. If the image is from a news event, search for the news event. If the image is from a person, search for the person. If the image is from a location, search for the location. Setting is the key. An image without setting is just an image. An image with setting is evidence.

The legal framework in Singapore is part of the setting, but it is not the verification method. The Protection from Online Falsehoods and Manipulation Act, or POFMA, is a statute that governs correction of online falsehoods. It is not a fact-checking agency. It is a law. The POFMA Office is part of the government, and it issues directions to correct falsehoods. A POFMA direction is not the same as a fact-check article, though both address falsehoods. The difference is that a POFMA direction has legal force. It requires the recipient to publish a correction. The correction must be approved by the government. The process is not an appeal. It is a direction. The recipient can appeal, but the appeal process is separate. The law is neutral in the sense that it applies to anyone who publishes a falsehood in Singapore, regardless of intent. The law is not the same as the truth. The law can correct a falsehood. It cannot make a claim true. The truth is determined by the evidence, not by the law.

POFMA and the Broadcasting Act

POFMA is Chapter 257A of the Singapore Statutes Online. The Broadcasting Act, Chapter 28, Part 10A, contains the online content powers. These powers allow the government to direct online services to block content that is against the public interest. The distinction between content moderation and legal action is the key. Platforms remove content under their terms of service. The government removes content under the law. The two are not the same. A platform may remove content because it violates the platform's rules. The government may remove content because it violates the law. The platform's rules are private contracts. The law is a public mandate. The difference matters because a platform can be arbitrary. The law is bound by due process. The law is also subject to review. The Protection from Harassment Act, or POHA, Chapter 256A, covers harassment and doxxing offences. The Penal Code covers other offences, including criminal defamation. The Personal Data Protection Act, Chapter 227A, covers data protection.

The Personal Data Protection Commission

The Personal Data Protection Commission, or PDPC, is the authority that administers the Personal Data Protection Act. The PDPC publishes enforcement decisions on data breaches. These decisions show organisations collecting, using or disclosing personal data beyond what is necessary or without adequate consent. The enforcement decisions are public. You can read them. They are a source of evidence about which organisations are trustworthy with your data. The actual data handling is a question of trust, not just of law. An app that claims to protect your data may not. The privacy policy is a contract. The enforcement decision is a fact. The privacy policy tells you what the organisation says it will do. The enforcement decision tells you what the organisation has actually done. The two may be different. The privacy policy is the intended behaviour. The enforcement decision is the actual behaviour. The discrepancy is the gap you should care about.

ScamShield and the Police

ScamShield is a specific tool. It is an app and a hotline that blocks scam calls and SMS and allows user reporting. It is not a general antivirus. It is a database of known scam numbers. The database is updated frequently. The update frequency matters because the scam numbers change daily. The false positive rate exists. A number that is not flagged is not proof of safety. The absence of a flag is not evidence of legitimacy. The Singapore Police Force publishes scam statistics. The National Crime Prevention Council runs an anti-scam website and a scam helpline. The Singapore Police Force's scam typology framework lists the main categories: job scams, phishing scams, e-commerce scams, investment scams, impersonation scams and love scams. The typology is useful because it helps you recognise the pattern. An impersonation scam poses as a government official, a bank officer or a family member. The goal is to extract money or data. The impersonation is the method. The fraud is the crime. The police are the first stop. The POFMA Office is the first stop for online falsehoods. The Personal Data Protection Commission is the first stop for data breaches. The order matters: the police for crimes, the POFMA Office for falsehoods, the PDPC for data.

The five-question routine is a habit, not a law. It does not make you immune to misinformation. It makes you slower. The slowness is the point. Verification speed matters because the first few hours after a claim appears are the most dangerous. During that window, the claim spreads, the screenshots are taken, and the correction, if it comes at all, arrives too late to undo the damage. The routine is designed to insert a pause between the impulse to share and the act of sharing. The pause is the space where the verification happens. The pause is the space where you decide. The pause is the space where you choose not to share.

The Failure Modes the Routine Prevents

Source confusion is the first failure mode. You believe a claim because the person who forwarded it is trusted, without checking whether that person is the original source. The routine prevents this by forcing you to identify the original source. Recency bias is the second failure mode. You believe a claim because it references a current event or uses fresh-looking screenshots, without checking whether the underlying information is old or fabricated. The routine prevents this by forcing you to check the setting. The verification bypass is the third failure mode. You share a claim with a comment like is this true? without waiting for an answer, amplifying the claim before verification. The routine prevents this by forcing you to wait. Tool over-trust is the fourth failure mode. You upload an image to a single AI-detection tool and treat the result as definitive, without understanding the tool's error rate. The routine prevents this by forcing you to use multiple tools and to understand their limitations. The officialdom halo is the fifth failure mode. You believe a piece of content because it uses a government logo, official-sounding language or a .gov.sg-like URL, without checking whether the communication channel is genuine. The routine prevents this by forcing you to verify the channel. Correction blindness is the sixth failure mode. You see a fact-check or correction but dismiss it because it comes from a source you distrust. The routine prevents this by forcing you to evaluate the evidence, not the source.

What the Routine Does Not Do

The routine does not replace the authorities. If you have been scammed, go directly to the Singapore Police Force, the POFMA Office or the Personal Data Protection Commission. The routine is for prevention. The authorities are for cure. The routine does not answer every question. It does not tell you whether a deepfake is real. It tells you to pause. It does not tell you whether an image is AI-generated. It tells you to search. It does not tell you whether a claim is true. It tells you to verify. The routine is a tool, not a solution. The solution is the habit. You build the habit by practising the routine on every claim, every image, every video, every piece of content. The first time you do it, it takes five minutes. The tenth time, it takes two. The hundredth time, it takes thirty seconds. The habit becomes automatic. The verification becomes instinct. The instinct is what you want. The instinct is the goal.

AI-content detectors are marketed as tools that can tell you whether a piece of text or an image was generated by artificial intelligence. The marketing is misleading. The real-world false positive rate is high. A false positive is when the tool says something is AI-generated when it was written or created by a human. The false positive rate matters because a false accusation is harmful. A student accused of using AI to write an essay may be punished for work they actually did. A journalist accused of using AI to write an article may be fired. The tools are not reliable enough to be used as evidence. The tools are a starting point, not a verdict. The user burden is real: each step costs time, and the total cost of verifying a single claim may be five minutes. The cost of sharing a falsehood is higher. The cost is your credibility, your relationships and potentially your money or your data. The tools are not the answer. The answer is the habit of verification.

The Specific Tools and Their Limits

ScamShield blocks scam calls and SMS using a known-scam database. The database is updated frequently. The update frequency matters because the scam numbers change daily. The false positive rate exists. A number that is not flagged is not proof of safety. The absence of a flag is not evidence of legitimacy. The same applies to AI-content detectors. The tools are not perfect. They have a real-world false positive rate. The false positive rate is the rate at which the tool says something is AI-generated when it is human-generated. The rate is not published, and it varies by tool. The tools are trained on a specific dataset, and the dataset may not reflect the real world. The tools are also subject to adversarial attacks. A person can intentionally modify an AI-generated image to avoid detection. A person can intentionally modify a human-generated image to trigger a false positive. The tools are not a reliable way to verify the origin of a piece of content.

What to Do Instead of Trusting a Single Tool

Instead of trusting a single tool, use multiple tools. Use a reverse image search. Use a reverse video search. Use a fact-checking site. Use your own judgment. The tools are a starting point, not a verdict. The verdict is the result of the full verification process. The process is the five-question routine. The routine is the habit. The habit is the goal. The tools are a means to an end. The end is the ability to judge whether a claim is true. The tools help you gather evidence. The evidence helps you make a judgment. The judgment is yours. The judgment is not the tool's. The tool is a tool. The tool is not a mind. The tool is not a replacement for critical thinking. The tool is a supplement to critical thinking. The tool is a starting point. The thinking is the process.

The problem with AI detectors is not just the false positive rate. The problem is the user burden. The user burden is the number of steps a person must take to verify a claim using a given method. The user burden for a reverse image search is low: you upload the image, and the tool gives you the results. The user burden for a deepfake detector is higher: you have to upload the video, wait for the analysis and then interpret the results. The user burden for a fact-checking website is higher still: you have to read the article, evaluate the evidence and decide whether the verdict is credible. The user burden is a real cost. The cost is time. The time is a scarce resource. The time you spend verifying one claim is time you cannot spend doing something else. The time is the price of the verification. The price is worth paying for a claim that could be a scam. The price is not worth paying for a claim that is trivial. The question is not whether you can verify a claim. The question is whether the claim is worth verifying.

The Verification Speed and the Update Frequency

Verification speed measures how quickly a fact-checking source publishes a verdict on a circulating claim. Update frequency measures how often a scam database or fact-check feed is refreshed. Both matter because the information ecosystem moves fast. A claim can go viral in an hour. A fact-check that takes three days to publish is useless. A scam database that is updated monthly will miss the scam that starts today. Verification speed and update frequency are the two numbers that tell you whether a source is useful. A source that publishes a verdict within hours and updates its database daily is worth checking. A source that publishes a verdict within weeks and updates its database monthly is not. Verification speed and update frequency are not the same as accuracy. A source can be fast and wrong. A source can be slow and right. Speed matters because the misinformation spreads fast. Accuracy matters because the correction must be right. The best source is both fast and accurate. The second-best source is accurate but slow. The worst source is fast and wrong. The worst source is the most dangerous.

The Actual Endorsement Problem

A piece of content may claim verification by the Singapore Police Force or the Ministry of Health. The claim may be false. The actual endorsement is a problem because the reader may believe the claim without checking. The officialdom halo is the tendency to believe content because it uses a government logo or official-sounding language. The halo is a bias, and the bias is exploited by scammers. Scammers use the logos because the logos work. The logos are a signal of trust. The trust is misplaced when the logo is a fake. The trust is well-placed when the logo is genuine. The problem is telling the difference. The difference is the channel. The genuine logo comes from the official website. The fake logo comes from a phishing email or a forwarded piece of content. The official website is the only source you can trust. The forwarded material is a source you cannot trust. The distinction is the first step. The second step is the verification. The verification is the process of checking the source. The process is the five-question routine.

The media literacy landscape in Singapore is composed of several institutions, each with a different role. The Media Literacy Council is a Singapore charity that runs public education campaigns on digital literacy. It is not a regulator. It does not enforce anything. It is not the same as the POFMA Office or the Personal Data Protection Commission. The distinction matters because a reader who confuses the two may expect the Council to act on a complaint, which it cannot do. The Council's role is education. The Council produces guides, runs workshops and partners with schools. The Council's work is important because it builds the skills that help people verify claims. The Council is not a verification tool. The verification tool is the routine. The Council is the teacher. The routine is the lesson.

The National Library Board and the S.U.R.E. Framework

The National Library Board is the agency that runs the public libraries. The Board has developed the S.U.R.E. framework as a teaching tool. The framework stands for Source, Understand, Research, Evaluate. It is a teachable sequence, not a technology. The framework is embedded in Singapore's school curriculum and public libraries. The framework does not depend on any specific technology. You can apply it to a printed newspaper, a broadcast segment or a forwarded piece of content. The framework is a habit. The habit is the goal. The framework is the structure. The habit is the practice. The practice is the verification.

The Personal Data Protection Commission and the PDPA

The Personal Data Protection Commission is the authority that administers the Personal Data Protection Act. The PDPA is the legislation that governs the collection, use and disclosure of personal data in Singapore. The PDPA's core obligations are consent, purpose limitation, notification, access and correction, protection, retention limitation, transfer limitation and accountability. These obligations matter because they give you rights. You have the right to ask what data an organisation holds about you. You have the right to correct it. You have the right to withdraw your consent. The PDPC publishes enforcement decisions on data breaches. These decisions are public. They are a source of evidence about which organisations are trustworthy with your data. The actual data handling is a question of trust, not just of law. An app that claims to protect your data may not. The privacy policy is a contract. The enforcement decision is a fact. The privacy policy tells you what the organisation says it will do. The enforcement decision tells you what the organisation has actually done. The two may be different. The privacy policy is the intended behaviour. The enforcement decision is the actual behaviour. The discrepancy is the gap you should care about.

Your digital footprint is the trail of data you leave through online activity, including posts, location data and browsing history. Your digital shadow is the data collected about you passively: browsing history, location tracking, data broker profiles. The distinction matters because the two have different implications for your privacy. Your footprint is something you actively leave. Your shadow is something that is collected about you. You can delete your footprint, at least in theory. You cannot delete your shadow, at least not easily. The shadow is collected by third parties who do not have a direct relationship with you. The shadow is bought and sold by data brokers. The shadow is used to target you with ads, to price-discriminate and to make decisions about your creditworthiness. The shadow is not something you can control. The shadow is something you can only mitigate. The mitigation is the same as the verification: check the privacy policy, limit the data you share, use the tools.

The Permanence of Digital Footprints

Content posted online may be archived, screenshotted or indexed even after deletion. This property of digital data is not changing. The permanence is a fact of life. The permanence matters because a claim you share today may be used against you tomorrow. Permanence is the reason you verify before you share. Permanence is the reason you think before you post. Permanence is the reason you treat every piece of content as if it were permanent. Permanence is a constraint. The constraint is the price of the convenience. The convenience is the ability to communicate instantly with anyone, anywhere. The price is the loss of privacy. The price is the loss of control. The price is the loss of the ability to take back what you have said. The price is worth paying for the benefits. But the price is real. The price is the reason the verification habit matters. The verification habit is the way you protect yourself from the permanence. It is the way you protect yourself from the consequences of sharing something false. It is the way you protect yourself from the scam. It is the way you protect yourself from the fraud. It is the way you protect yourself.

The final challenge is the language barrier. A claim arrives in a language you do not speak, forwarded with a translated summary. The summary may be accurate. The summary may be a lie. The summary is a translation, and a translation is an interpretation. The interpretation may be wrong. Verification of a claim in a foreign language follows the same five-question routine, but with an extra step. The extra step is the translation. Find a second translation. Compare the two. Look for discrepancies. Check the source. Check the evidence. Check the setting. Check the intent. Check the emotion. Emotion is the hard part. A piece of content in a foreign language may not provoke the same emotion in you as it does in a native speaker. Emotion is the signal. The signal may be lost in translation. The signal matters because emotion is the shortcut the scammer uses. The scammer wants you to feel the emotion before you think. The scammer wants you to act on the emotion. The verification habit is the way you resist. The verification habit is the way you pause. The verification habit is the way you think.

The Translated Summary Trap

A translated summary is a second-hand account. The summary is not the original. The summary is a version. The version may be accurate. The version may be inaccurate. The version may be a lie. The lie is the danger. The lie is the trap. The trap is the content that tells you what you want to hear. The trap is the content that confirms your bias. The trap is the content that makes you feel good. The trap is the content that makes you feel angry. The trap is the content that makes you feel afraid. The trap is the content that makes you feel hopeful. The trap is the content that makes you feel anything. The feeling is the trap. The trap is the emotion. The emotion is the shortcut. The shortcut is the way around your critical thinking. The shortcut is the way around your verification habit. The shortcut is the way around the routine. The routine is the protection. The routine is the guard. The routine is the gate. The gate is the way you stop the trap. The gate is the way you stop the shortcut. The gate is the way you stop the emotion. The gate is the way you stop the share. The gate is the way you stop the scam.