The basic idea
AI can make deception easier to create and harder to recognise
Artificial intelligence can generate convincing text, images, audio and video. These capabilities have many legitimate uses, but they can also be misused to create fraudulent messages, impersonate people or spread false and misleading information.
A scammer no longer needs perfect writing skills or advanced media-production experience to create professional-looking content. AI tools can help produce polished messages, imitate communication styles and create synthetic media quickly.
This does not mean that everything online should be treated as fake. It means that appearance alone is becoming less reliable as proof of authenticity.
AI-enabled scams
Familiar scams can become more convincing with AI
Many AI-assisted scams use familiar fraud techniques. The difference is that AI can help criminals create more believable messages, personalised stories, fake documents or convincing impersonations.
Unexpected urgency
The message pressures you to act immediately, send money, reveal information or click a link before you have time to think.
Requests for money or credentials
Be suspicious when someone unexpectedly asks for passwords, verification codes, card details, bank information, cryptocurrency, gift cards or urgent transfers.
Unusual communication
A familiar person may contact you from a new number, account or email address, or communicate in a way that does not match their normal behaviour.
Emotional pressure
Scammers may create fear, excitement, sympathy or panic so that you react emotionally instead of verifying the situation.
Suspicious links or attachments
Links may imitate legitimate websites, while attachments can be used to steal information or deliver malicious software.
Requests for secrecy
A scammer may tell you not to contact family members, colleagues, banks or other people who could help you verify the story.
Example
A convincing emergency message
Imagine receiving a message that appears to come from a relative, colleague or manager. It says there is an emergency and asks you to transfer money immediately. The wording sounds natural and may even include personal details.
Instead of trusting the message because it sounds convincing, contact the person using a phone number or communication method you already know. Independent verification can stop a believable impersonation from becoming a successful scam.
Deepfakes
What is a deepfake?
A deepfake is synthetic or manipulated media that can make a person appear to say or do something that did not happen as presented. Deepfakes can involve video, images or audio.
Some synthetic media is created for entertainment, education or creative work. The danger comes when manipulated content is presented deceptively to impersonate someone, damage a reputation, commit fraud or mislead an audience.
When something important looks or sounds real, verify it
- Consider whether the video, image or audio came from a trustworthy original source.
- Look for confirmation from official accounts, reputable news organisations or other independent sources.
- Be cautious when sensational content appears only on one unfamiliar account or website.
- Do not assume that a realistic voice, face or video automatically proves that a person really said or did something.
- For important requests, verify the person's identity through a separate communication channel you already trust.
Misinformation
AI-generated content can spread incorrect information at scale
Misinformation is false or inaccurate information that can be shared regardless of whether the person spreading it intends to deceive others. Deliberately deceptive information is often described as disinformation.
Generative AI can produce large amounts of convincing content quickly. If incorrect claims are copied, reposted or repeated without verification, they can travel across social media, messaging platforms and websites rapidly.
Before accepting or sharing an important claim, ask where it came from, whether the original source can be found and whether reliable independent sources support it.
A practical verification habit
Stop, check and verify before you act
Stop
Do not immediately click, pay, forward, download or provide information.
Check the source
Examine who sent or published the information and whether the account, address or website is genuine.
Verify independently
Use another trusted source or communication channel instead of relying on the contact details supplied in the suspicious message.
Look for supporting evidence
Search for reliable confirmation, original statements, official announcements or reporting from multiple credible sources.
Think before sharing
If you cannot establish that the information is genuine, avoid spreading it to other people.
Important
Do not rely on a single visual clue to detect a deepfake
Synthetic media technology changes quickly. A strange facial movement, unusual voice or visual mistake may raise suspicion, but the absence of obvious mistakes does not prove that content is genuine. Source verification and independent confirmation are stronger habits than relying only on appearance.
Protecting yourself
Verification becomes more important as generated content improves
If a message asks for money, sensitive information or an important action, verify the request using contact information you already trust. Do not use a phone number or link supplied only by the suspicious message.
For organisations, clear approval procedures can also reduce risk. Important payments, account changes and sensitive requests should not depend only on a voice message, email, video call or chat message appearing to come from a familiar person.
Key takeaways
What to remember
- AI can make fraudulent messages, images, audio and videos more convincing.
- A realistic-looking piece of content is not automatically authentic.
- Urgency, secrecy and unexpected requests for money or sensitive information are important warning signs.
- Verify important requests through a separate trusted channel.
- Check claims against reliable and independent sources before sharing them.
- Human judgement and verification remain essential even as AI-generated content becomes more realistic.
Next lesson
What AI Can and Cannot Do
In the next lesson, we will look more closely at the strengths and limitations of artificial intelligence so you can develop realistic expectations about where AI is useful and where human expertise remains essential.