When to Teach Adults Age 47 Vocabulary Words

When should you teach vocabulary words to an adult aged 47? Teach them before someone trusts a voice message, shares an apparently authentic video, follows a request from a supposed official or transfers money after an urgent call. Terms such as deepfake, synthetic media, voice cloning, impersonation, verification, source authentication, phishing, spoofing, fabricated evidence and independent channel are useful because today’s convincing voices and faces are not always proof of who produced a message. Explain the words at the moment they protect a real choice, then practise verifying without clicking suspicious links.

Age forty-seven is not a special vulnerability threshold. Deepfake scams affect people of many ages, and digital familiarity does not make someone immune. Earlier guides in this series cover reading research claims and general account security; this chapter focuses on the authenticity of voices, videos and apparent authority. The goal is careful verification, not fear of all technology or assumptions that every edited image is malicious.

Quick answer: when the vocabulary should be taught

When a familiar voice requests money: teach voice cloning, impersonation and callback verification before responding.
When a celebrity or official appears to endorse an investment: identify deepfake, synthetic media, claim, original source and verify through independent official channels.
When a video meeting contains a sudden payment request: distinguish identity assurance from a face merely appearing on screen.
When a message uses an official-looking sender: explain spoofing, sender identifier and independent verification.
When a video has odd visual details: use possible manipulation rather than declaring authenticity from one clue.
When a relative receives an alarming call: understand urgency, coercion, pretext and how to check the story safely.
When a scam-checking tool returns a warning or uncertain result: distinguish risk indicator from a guaranteed verdict.
Before forwarding a dramatic clip: pause, check its context and identify whether the source is genuine.

The key idea: the message may look real, but identity still needs verification

A deepfake is an image, video or audio recording created or manipulated using synthetic methods to convincingly represent something that did not occur in that form. It can be used harmlessly in entertainment, but scammers can misuse it to appear to be a manager, relative, financial authority or well-known person. The important mistake is treating a recognisable face or voice as sufficient identity evidence.

Singapore’s Cyber Security Agency (CSA) advisory on deepfake scams describes an approach using three groups of checks: assess the message, analyse the audiovisual elements and authenticate using relevant tools. The CSA warns that deepfakes may be extremely realistic. The absence of odd blinking or visible glitches does not prove that a message is genuine.

The Ministry of Home Affairs’ August 2026 answer on real-time deepfake impersonations describes government efforts and stresses checking purported government requests through reliable channels. It also explains that government officials will not ask people over the phone to transfer money, disclose bank login details or install apps from unofficial stores.

A useful vocabulary question is: “Who can independently confirm this request?” A video meeting, caller ID, an account photo and a copied logo are clues, not conclusive authorisation.

An ordinary-language checklist before responding

Stop the requested action: do not send money, codes or confidential information while identity remains uncertain.
Identify the claim: who does the caller say they are, and what exactly do they want?
Name the possible mechanism: voice cloning, impersonation, spoofed sender, manipulated video or a simpler fraud.
Find an independent contact: use a number or official app you already trust, not the contact details supplied in the suspicious message.
Confirm the purpose: ask the actual organisation or person whether they requested this action.
Escalate appropriately: if you suspect a scam, use official scam-prevention or reporting channels and your financial institution if payment is involved.
Do not forward unchecked content: avoid spreading a clip whose origin is not known.

If a transfer has already occurred, act promptly through the bank’s official fraud process and relevant local authorities. This article does not replace incident-response assistance.

Thirty important AI authenticity and scam words

Artificial intelligence (AI): computer methods for tasks such as generation, classification or prediction.
Generative AI: systems that create content such as text, images, audio or video.
Synthetic media: media content produced or modified through digital generation.
Deepfake: convincing synthetic or manipulated media representing a person or event.
Voice cloning: generating speech resembling a particular person’s voice.
Impersonation: pretending to be someone else.
Spoofing: making an identifier or communication appear to come from a different source.
Caller ID: the displayed information about a calling party, which is not always proof of identity.
Sender ID: a displayed label associated with a message source.
Phishing: deception intended to make people reveal information or take unsafe actions.
Pretext: a story used to justify a request, potentially falsely.
Social engineering: manipulating human decisions to obtain information or access.
Urgency cue: language used to create pressure for immediate action.
Coercion: pressure or threats used to force action.
Authenticity: whether content or identity is what it claims to be.
Verification: checking information against trustworthy independent evidence.
Authentication: confirming identity or access under an appropriate system.
Original source: the first or authoritative source for the claimed message.
Independent channel: a separate, trusted communication route used to confirm a claim.
Callback verification: contacting a person through a pre-existing trusted number rather than the caller’s supplied details.
Manipulated audio: altered sound, which may or may not involve AI.
Manipulated video: altered moving images that change apparent content.
Fabricated evidence: material created to make a false claim appear supported.
Context: surrounding facts needed to interpret a message.
Risk indicator: a sign of possible danger, not necessarily conclusive proof.
False positive: an innocent item mistakenly flagged by a detection process.
False negative: a harmful item not identified by a detection process.
ScamShield: Singapore’s set of scam-checking and prevention tools.
Official channel: a communication path verified as belonging to the responsible institution.
Recovery: steps to regain control or address damage after an incident.
Report: notifying the responsible service or authority with relevant information.

Words that sound technical should be used accurately. Not every misleading video is a deepfake; some are simply old footage, selective edits or genuine footage with a false caption. Detecting a manipulation and understanding a claim are related but separate questions.

Nine realistic situations and what to teach

1. A relative’s voice says, ‘I need money urgently’

A family member receives a voice message that sounds like someone they know. It reports an emergency and demands immediate payment to a new account. This is the moment to teach voice cloning, pretext and callback verification. The listener should reach the relative through an independently known number or another trusted method before acting.

Do not assume a familiar accent, nickname or emotional delivery confirms identity. Scammers may know personal details from previous contact or public social media. A real emergency should be verified through trusted relevant services rather than a new account number supplied by a stranger.

2. A public figure appears in an investment advertisement

A video shows a recognisable public figure endorsing an unusual scheme. Teach synthetic media, endorsement and source verification. Check the person’s verified channels and relevant financial regulator before concluding that the claimed endorsement exists. Even an authentic endorsement would not prove that an investment is safe or suitable.

CSA’s deepfake advisory includes examples of impersonation used in financial scams. Do not click a promotional payment link just to see whether it is authentic.

3. A supposed manager appears on a video call demanding a transfer

A staff member sees what looks like the organisation’s manager or finance director on screen. An urgent request for a new transfer appears. Teach transaction authorisation, out-of-band verification and dual approval. The staff member should follow the organisation’s established payment-control process independently of what appears on the call.

CSA’s advisory discusses a case in which highly convincing deepfake impersonation contributed to a large financial scam. The lesson is not to distrust every manager, but to avoid allowing a single simulated meeting to bypass financial safeguards.

4. A strange blink or mismatched lip movement appears

An observer sees unnatural facial movement in a video and concludes “This is definitely fake.” Teach possible manipulation, artefact and uncertainty. Blurred edges or imperfect synchronisation can raise suspicion, but network compression, lighting or poor video quality can also produce odd effects.

A carefully generated fake may have none of the familiar glitches. The CSA’s guidance treats visual analysis as one step alongside source checks; visible clues alone do not deliver certainty.

5. A message carries an official-looking government identity

An SMS has a familiar sender label, but its content asks for credentials or a transfer. Teach sender identifier, official communication and independent verification. Verify the specific demand through the actual government service, not simply by trusting a displayed label or following a link.

Singapore’s gov.sg guidance on digitally altered content discusses verified official websites and government SMS identification. Scammers may still imitate surrounding content or use non-SMS channels; assess the actual request.

6. A video claims an emergency has occurred nearby

A dramatic clip shows an alleged incident but does not provide a date, original location or trustworthy source. Teach context, unverified and misattribution. The footage may be real but from another time or place, synthetic, or selectively presented. Check relevant authoritative news and emergency sources before forwarding.

Do not tell people to ignore an immediate emergency they can directly observe. Safety comes first. For an online claim about a distant incident, independent verification is the appropriate response.

7. An automated scam checker returns an uncertain result

A checker may flag a suspicious URL or report that there is insufficient information. Explain risk indicator, false positive and false negative. A positive warning warrants care, but a clean or uncertain result does not guarantee safety. Context and independent verification remain important.

Singapore’s ScamShield check-for-scams page explains using its app or calling 1799 for help assessing suspicious messages, numbers and links. The service supports checking; it does not replace responsible decisions about sharing money or credentials.

8. Someone has already shared a verification code

If a person realises they gave a code to an impersonator, explain credential compromise and account recovery. Act through the real service’s recovery procedure, secure relevant accounts and contact the bank’s fraud team if financial details were involved.

Avoid posting screenshots of personal login codes or payment records online. ScamShield and local law-enforcement resources can help locate the correct reporting route, but the affected service’s official instructions control account recovery.

9. The family wants to create a ‘secret phrase’

A pre-agreed family phrase may be a supplemental check in some situations, but a phrase can be discovered, shared or coerced. Teach authentication factor and independent confirmation. A call back through a known number, contact with another trusted person or a verified authority may be more reliable than treating one shared phrase as an unbreakable password.

No simple family technique guarantees protection. The emphasis should remain on pausing unusual financial or information requests and verifying through separate, trusted channels.

Three worked comparisons

An authentic voice versus an authorised request

Scenario A: a familiar voice requests a transfer to an unfamiliar account. Scenario B: the person independently contacts the relative through a previously known channel and confirms no such request was made. The apparent voice was not enough to establish identity or authorisation. A request must be checked separately.

A video with glitches versus a convincing fake

Video A: lip movements appear slightly out of sync. Video B: everything looks natural, but no original source is available and the video makes an extraordinary claim. Neither observation proves authenticity. Verify the claimed event through authoritative sources and relevant participants rather than relying solely on visual anomalies.

An official-looking text versus a verified instruction

Text: “Immediate payment required or your account will be suspended. Click here.” Safer response: open the known official service independently and check the account status or call a verified number. The critical words are urgent request, impersonation, official source and verification.

A four-week practical vocabulary plan

Week 1 — Learn the mechanisms: distinguish deepfake, voice cloning, spoofing, phishing and ordinary misleading captions using harmless fictional examples.
Week 2 — Understand the evidence: compare authenticity, original source, verification, visual artefact and uncertainty without assuming appearances prove identity.
Week 3 — Practise safe responses: rehearse how to decline urgent requests, contact a known person independently and locate official scam resources.
Week 4 — Apply the language: assess an invented suspicious message, separate observation from assumption and write the safest next question.

The four-week plan is a flexible learning structure, not a proven safety course. Practise on fictional media, not with real stolen recordings, passwords or unauthorised impersonations.

How to tell the word has become useful

Someone can define deepfake but still transfer money because a video looked real. Deeper understanding appears when they pause, verify through an independent channel and follow payment controls. Someone who knows phishing should be able to recognise that a link’s request for credentials is a reason for caution even if no AI-generated video is involved.

Use a new fictional scenario each time. The learner should identify the apparent claim, the risk, the missing evidence and the safest method of checking. Accuracy and calm action matter more than diagnosing every clip.

Common errors to correct

Assuming every edited clip was made by AI: editing, miscaptioning and synthesis are different mechanisms.
Treating natural-looking footage as proof: realistic deepfakes may lack obvious visual errors.
Trusting caller ID as identity confirmation: displayed identifiers can be spoofed or misleading.
Acting under urgency before checking: unusual requests deserve separate verification.
Believing a scam-checker guarantees safety: automated indicators have limits.
Forwarding alleged evidence before its source is known: this can spread harm and misinformation.
Sharing recovery codes with helpers: use only official account recovery channels.

Frequently asked questions

What is a deepfake?

It is synthetic or manipulated media that convincingly represents an event or person. Deepfakes can have legitimate uses, but they can also support deception and impersonation.

Can I always detect a deepfake by watching the eyes and lips?

No. Visual clues can be helpful but are not reliable proof. Confirm the source and request through independent, appropriate channels.

Is voice cloning the same as caller-ID spoofing?

No. Voice cloning imitates how someone sounds; caller-ID spoofing changes or misrepresents the displayed origin of a call. They can occur separately or together.

Can ScamShield tell me with certainty whether a message is safe?

It is a useful checking and support resource, but any single tool has limitations. A suspicious request for money or credentials should be independently verified.

What should I do when a known person seems to request urgent money?

Contact that person through a separately known channel and do not transfer funds or share codes until the request is genuinely verified.

Why focus on age 47?

The age is an editorial navigation marker, not a claim that one age group is uniquely vulnerable to deepfakes or scams.

Where can I find current Singapore scam help?

The official ScamShield website provides resources, its app and the 1799 helpline. For a completed financial transfer, also contact the affected bank through official channels promptly.

Official references and related eduKate learning

CSA — Detecting and Responding to Deepfake Scams; MHA — Real-Time Deepfake Impersonation Measures (August 2026); ScamShield — Check for Scams; gov.sg — Digitally Altered Content; and National Crime Prevention Council — ScamShield.

For passwords and sharing permissions, see Age 27 — Digital Security Vocabulary. The previous guide is Age 46 — CPF and Retirement Vocabulary. Continue through Vocabulary by Age and Master Vocabulary.

Bottom line: At forty-seven—or any age—teach words such as deepfake, impersonation and verification when a convincing voice, video or identity claim might influence real action. Verify the source and request independently before transferring money or sharing private information.

When to Teach Vocabulary | Related Adult Age Guides

Age 46 guide · Age 48 guide · Vocabulary by Age series hub · Master Vocabulary. Adult ages are navigation labels, not compulsory milestones or vocabulary targets.