Criminals are increasingly weaponising artificial intelligence to create convincing fake websites, clone voices and generate deepfake videos designed to steal money from unsuspecting shoppers and online users. According to BBC News, fraudsters are deploying these tools at scale to supercharge traditional scams, making them harder than ever to detect. For the first time, National Trading Standards has published an AI scam watchlist to help consumers understand the techniques criminals are using and how to defend themselves.
The sophistication of AI-powered fraud is creating widespread concern among the public. A Censuswide survey of 2,000 UK adults found that 47% said it was getting harder to tell what was genuine online, 24% felt overwhelmed by scam sophistication, and 19% said they avoided some online activities because of scam concerns. This anxiety reflects a real and growing threat: UK fraud cases have surged as criminals use AI to manipulate victims, with losses nearing £1.3bn in 2025, according to UK Finance.
What scam tactics are criminals using?
Fraudsters are deploying AI in multiple ways to deceive consumers. Fake retail websites are being created with professional-looking designs that claim to be family-owned UK businesses selling quality homemade products. These sites often feature AI-generated images showing British High Streets complete with Union Jack flags. When customers place orders, they either receive nothing or get cheap, mass-produced items shipped from countries such as China, trading standards officers warn.
Romance scams represent another major threat amplified by AI technology. Criminals use artificial intelligence to remember victims' personal details and sustain conversations over weeks or months, building false emotional connections before requesting money. One case involved a daughter's account of her mother losing £140,000 to a romance scammer using deepfake technology and emotional manipulation.
Deepfake videos and voice-cloned audio are being weaponised to impersonate family members or close friends, with scammers claiming an emergency requires immediate money transfers. Meanwhile, phishing campaigns have become far more effective: AI can now generate error-free, personalised emails, text messages and social media posts at enormous scale, eliminating the typos and grammatical mistakes that once helped consumers spot fraudulent messages.
On 9 June 2026, National Trading Standards warned of a particularly insidious tactic: criminals were manipulating AI models to recommend scam websites in search results, with the sites reportedly attracting hundreds of visitors a day through AI recommendations. The watchlist also identifies poisoned AI search results, AI chatbots generating fake reviews, and compromised recommendation systems among the techniques in criminals' toolkit. Scam types being amplified include travel and ticket fraud, technical support scams, fake charity appeals, rogue trader schemes and recovery fraud.
Why are traditional warning signs no longer reliable?
Consumers have long been taught to spot scams by looking for telltale signs of poor quality: spelling mistakes, grammatical errors and amateurish design. These defences are now obsolete. Consumer group Which? points out that AI can produce error-free communications indistinguishable from legitimate businesses, making it impossible to rely on sloppy writing as a warning signal. National Trading Standards notes that AI is generally making existing scams more convincing and scalable rather than creating entirely new scam types, but warns that future AI agents could carry out complex fraud with minimal human involvement.
How can you protect yourself from AI scams?
Trading standards recommend several practical steps to reduce your risk. Never send money, gift cards, or personal or financial information to someone you have not met in person, no matter how convincing they seem online. Conduct thorough research before making purchases or financial commitments: verify whether a shop's address is real by checking it independently, use the Financial Conduct Authority's Firm Checker to confirm a company is authorised to give financial advice, and contact businesses directly to verify job adverts are genuine.
Do not assume someone is genuine simply because you have heard their voice or seen them on a video call. Deepfake technology has advanced to the point where audio and video can be fabricated convincingly. Instead, adopt a cautious approach to any request for money or personal information: stop and think before paying, sharing details or taking action. Be especially wary of pressure or urgency, as criminals deliberately create time pressure to prevent you from thinking clearly and verifying claims.
If you believe you have been targeted by a scam, report it immediately. Fraud and scams can be reported to Report Fraud or by calling 0300 1232040. For advice and support, contact the Citizens Advice Consumer Helpline on 0808 223 1133.
What does the broader picture show?
The rise of AI-powered scams reflects a fundamental shift in how criminals operate. Rather than relying on individual skill or effort, they can now automate and scale their attacks across thousands of potential victims simultaneously. Recent cases have shown shoppers losing hundreds of pounds after buying from fake websites that claimed to be Welsh businesses but were later found to use stolen images. The combination of AI-generated content, automated targeting and emotional manipulation creates a formidable threat that requires consumers to be more vigilant than ever before.
