What Happens When Scammers Start Sounding Real?

Vishing—short for “voice phishing”—has evolved. With artificial intelligence, AI voice phishing scams now sound frighteningly real, mimicking your bank, your boss, or even your business partner.

Using AI-generated voices and even deepfake video calls, attackers are making social engineering more persuasive—and harder to spot. As a result, detecting these scams has become far more challenging.

This isn’t science fiction. These attacks are happening right now.

AI-Powered Voice Scams

Attackers are using AI to:

• Clone the voices of executives or colleagues
• Create realistic helpdesk-style chatbot interactions
• AI voice phishing attacks often use cloned voices to pressure employees into action
• Conduct live “voice clone” calls pretending to be senior staff

Attackers often use these tactics to:

• Push through unauthorised payments
• Trick staff into revealing login credentials
• Initiate fraud during financial transactions

Real example: A finance worker in London authorised a £25.6 million transfer after joining a deepfake video call impersonating their CFO and other staff.

Vishing: Old Scam, New Tech

In contrast, traditional voice phishing (vishing) involves scammers calling you and pretending to be:

• A bank or government official
• A colleague from IT or HR
• A supplier needing urgent details

Now, with AI:

• Caller ID spoofing makes calls look legitimate
• Voice synthesis makes the scammer sound like someone you know
• Scripted urgency forces snap decisions

Common red flags:

• Pressure to act fast
• Fear of consequences (“your account will be locked”)
• Instructions to override normal procedures
• Requests for codes, logins, or payment details

Deepfakes: Now in Your Inbox (or Meeting Room)

Similarly, deepfake videos use AI to mimic a real person’s face and voice. Scammers have used them to:

• Impersonate CEOs giving urgent instructions
• Trick employees during live video calls
• Launch fake investment or financial requests
• Spread disinformation across social media

Signs it’s a deepfake:

• Lip movements not matching speech
• Strange eye behaviour or expression glitches
• Robotic or unusual tone
• Overly urgent or uncharacteristic requests

For more on how AI is reshaping phishing, read the National Cyber Security Centre’s guidance on deepfake and voice scams.

How to Stay Safe from AI-Based Scams

1. Pause and verify
If a request seems urgent or out of character—don’t act on it without checking.

2. Use a separate channel
Call, message, or video chat via a method you trust—not the one the scam came through.

3. Educate your team
Ensure staff are trained to spot voice manipulation and phishing red flags.

4. Don’t share personal voice/video content unnecessarily
Public videos or audio clips can be scraped and cloned. Limit what’s shared online.

5. Report everything suspicious
Even if you’re unsure, flag it. It helps protect others on your team.

Final Thoughts: Don’t Trust the Voice—Trust the Process

Scammers are no longer just trying to outsmart your systems—they’re trying to impersonate your people. And when the voice sounds familiar, it’s even easier to be caught off guard.

Above all, stay calm—and confirm before you comply. Aalways trust your gut if something feels off. The best defence against AI voice phishing is verification before action.

Want Help Defending Against AI-Era Scams?

Novo IT provides cyber awareness training, phishing simulations and domain-level protection against spoofing and impersonation—so your people stay protected even when the threats sound all too real.

In December, we come back to basics—because good data habits still beat the best scam.

👉 Explore our Cybersecurity Services or reach out to us at: 📧 [email protected] | 📞 01293 664413

Disclaimer: This article was created with the assistance of AI tools based on prompts and guidance provided by Novo IT Ltd.

---