The centrality of social media to the massive growth of fraud in the UK poses questions as to whether fraud investigators have all the technological tools they need. Many scams and Ponzi schemes target victims or recruit accomplices on social media platforms where informal networks also share methods relating to retail, insurance, benefits or public sector fraud.
It is an escalating problem. The ONS Crime Survey for example, believes fraud has reached unprecedented levels with an estimated 4.1 million incidents in 2024 across England and Wales – a 33% increase year-on-year.
Consumer and retail fraud increased by 35%. UK Finance estimates criminals stole £1.2bn in unauthorised and authorised fraud in 2024. In April this year, the financial institution Barclays’ Scams Bulletin, reported a 20% year-on-year increase in such frauds over the quarter, with losses per victim averaging £8,000 (£19,000 among the over-60s).
Social media and dark web analysis at scale is critical to the detection and prevention of schemes such as romance fraud, where criminals use familiar platforms as well as dating apps. The “Tinder Swindler,” Shimon Hayut, who was arrested in Georgia in September 2025, had previously managed to combine romance fraud with a multi-million-pound Ponzi scheme.
The range of scams on social media is constantly expanding – from fake adverts to phoney legal compensation or refund schemes, ticket fraud, or advice on how to tamper with energy meters. The start of the new university year saw many social media scams targeting students living away from home for the first time.
One of the mechanisms behind last year’s surge in unauthorised payment fraud (3.13 million cases, up 14% on 2023) was the use of social media to obtain bank or card details, often through the sale of cheap goods.
AI-powered social media intelligence tools and criminal connections
Automated social media data collection and AI-enabled analysis are vital for exposing suspected criminals and their networks, as the amount of information available is beyond human capabilities.
With so much fraud being initiated online, fraud detection agencies, including constabularies, the National Crime Agency and City of London Police, along with the Insurance Fraud Bureau, could all benefit from advanced, AI-powered open-source intelligence (OSINT) tools.
Automated social media data collection and AI-enabled analysis are vital for exposing suspected criminals and their networks, as the amount of information available is beyond human capabilities. Statista, the statistics website, estimates there are nearly 55 million UK social media users. It expects the number to hit 65 million by 2027. Finding the right account in this giant haystack of data is impossible through manual processes alone. Even with the best will in the world, it is unlikely that investigators can keep pace, not just with the volume of data, but also with the creation and use of new platforms.
It is simply impossible for an analyst or officer to monitor and investigate all the mainstream platforms and explore the obscurities of the deep and dark web where criminals collaborate, exchange ideas and recruit. They need to combine their investigative skills with OSINT technology that automates data collection, quickly filters and correlates massive amounts of multimedia data to detect risks, and integrates external intelligence sources. This is the solution to tackling fraud – not simply increasing headcount or refining outdated manual processes.
Fraud-as-a-Service detection
It is the hidden networks that AI-powered OSINT technology can expose through analytics and technologies such as OCR (optical character recognition).
Take the example of Fraud-as-a-Service – a US phenomenon that has now spread to the UK, enabling inexperienced individuals to execute fraud through phishing kits, account takeovers and other methods. Many of the perpetrators are in their teens and use platforms such as Discord to plan and share plans, tips and tricks for carrying out fraudulent acts.
In July this year, the student Ollie Holman was jailed for seven years for creating and supplying more than a thousand phishing kits that mimicked charity, bank and governmental websites. He may have enabled up to £100m in fraud. His network of connections was 700-strong, and the Crown Prosecution Service emphasised that he had been unable to conceal his online activities, despite using encrypted messages on Telegram.
It is the hidden networks that AI-powered OSINT technology can expose through analytics and technologies such as OCR (optical character recognition). These OSINT tools are easy to use and deliver fast and accurate insights, enabling investigators to make informed judgments.
Technology in combination with human expertise and decision-making
Purpose-built for uncovering threats across the surface, deep and dark web, OSINT solutions can establish indicators of fraud and map out connections and networks very rapidly. Combined with other forms of human and traditional intelligence, OSINT technology provides vital insights, going beyond keywords to assist in analysing images, videos, memes, posts and other forms of textual content. If technology cannot reveal the main entity behind content, associates with online links often can be identified through the power of analytics.
Request Fivecast’s industry brief to learn more.
The mission of Fivecast is to enable a safer world. As a world-leading provider of digital intelligence solutions, Fivecast helps the world’s most important public and private organizations explore masses of data, uncovering actionable insights which are critical to protecting global communities. Purpose-built to address the highest priority use cases in the national security, law enforcement, defence, corporate security and financial intelligence markets, Fivecast deploys advanced data collection and AI-enabled analytics to solve the most complex intelligence challenges.
Visit Fivecast to learn about their innovative solutions and how they are helping law enforcement agencies enhance their investigations.
