AI makes fake support scams harder to spot

• Generative AI is powering more sophisticated and realistic scams
• Fraudsters use AI to mimic brands and clone voices for deception
• Users must verify accounts and avoid sharing sensitive data via chat

You tweet at an airline about a delayed flight and instantly get a reply from a helpful-looking support account. But as you follow their instructions, you realise it’s a scam. This scenario is becoming more common and harder to avoid.

The Old Scams Get a Major Upgrade

For years, the advice for spotting online fraud was simple: look for typos, bad grammar, and suspicious links. Scammers operating from overseas often gave themselves away with poorly written messages. But generative artificial intelligence has erased those tell-tale signs. Today, criminals use AI to create flawless, context-aware, and highly personalised messages at a massive scale. These tools can scrape social media for details about your situation, allowing them to craft replies that seem genuinely helpful and specific to your problem. What used to take hours of manual effort can now be automated, with AI generating convincing text, realistic profile photos, and even entire fake websites in minutes. The result is a new breed of fraud that is more sophisticated and successful than ever before.

How AI Is Powering the Perfect Scam

Artificial intelligence gives scammers a powerful toolkit. They can create AI-powered chatbots that convincingly mimic the tone and language of a real company's customer service team. These bots can be deployed on fake websites that are nearly identical to the real ones, tricking users into handing over sensitive information like passwords or bank details. Scammers also seed the internet with fake customer support phone numbers on blogs and forums, knowing that AI-powered search summarisation tools might scrape this false information and present it as a legitimate contact number. In some cases, fraudsters use AI to clone voices from audio clips found online, allowing them to impersonate a company representative or even a family member in distress over the phone with terrifying accuracy.

Red Flags in the Age of AI

While AI makes scams slicker, some classic red flags remain. The biggest is a request to move the conversation to another platform, like WhatsApp or Telegram, away from the public eye of social media. Another is any request for personal or financial information, such as your password, OTP, or credit card number. Legitimate companies will never ask for this information in a chat or direct message. Scammers also rely on creating a sense of urgency, pressuring you to act quickly before you have time to think. You should also be wary of accounts with very recent join dates, low follower counts, or handles that are slightly misspelled versions of official brand accounts. Even if an account has a verified checkmark, proceed with caution, as verification can sometimes be bought.

Your Five-Step Defence Strategy

Protecting yourself requires a new level of vigilance. First, always verify an account before engaging. Check the company's official website for their designated social media handles. Second, never share sensitive data. No legitimate support agent needs your password or full bank account number to help you. Third, avoid clicking on links sent in unsolicited messages or texts; instead, navigate to the company's website or app yourself. Fourth, if a conversation feels 'off' or the 'agent' is pressuring you, stop and disengage. Finally, if you encounter a suspicious account, report it to the social media platform immediately to help protect others. The core principle is to treat any unexpected offer of help with healthy scepticism and always default to official channels you've verified yourself.