In recent years, telecom and online fraud cases have become increasingly common, with ever-changing tactics spreading across the Mainland, Hong Kong and Taiwan. These scams range from investment fraud and impersonation of officials to online shopping traps. The rising number of cases and total financial losses have drawn wide public concern.
To better understand how citizens in the three regions behave and perceive risks when faced with fraud, a research team led by Professor Christine Huang Yi-hui, Head and Chair Professor of the Department of Media and Communication in the College of Liberal Arts and Social Sciences of City University of Hong Kong (CityUHK), conducted a large-scale cross-regional survey on telecom and online fraud.
The study found that in both Hong Kong and the Mainland, victims were predominantly highly educated (with a bachelor's degree or above). It also revealed that victims in Hong Kong suffered the most severe financial losses among the three regions.
Covering over 4,000 respondents, the research sheds light on the diverse risk landscapes of online fraud in different areas and provides valuable references for anti-scam education and policymaking.
The survey revealed that more than 95% of respondents had encountered scam messages, with similar a proportion in the Mainland (96.2%), Hong Kong (97.4%) and Taiwan (96.5%). However, among those who reported financial losses, Hong Kong victims suffered the most severe losses, with nearly 55% (54.9%) losing over HK$100,000 - far higher than the number in Taiwan (18.9%) and the Mainland (39.7%).
Victim profiles also varied across the three regions. In Hong Kong, individuals with postgraduate qualifications were more likely to respond to scam messages and incur financial losses. Women and those who interacted with scammers for three to six months suffered the highest losses. In Taiwan, women over the age of 40 recorded greater financial losses than other groups. In the Mainland, victims with a bachelor's degree and those aged over 50 were more likely to reply to scam messages and incur losses, with the 60+ age group experiencing the largest financial losses. Longer interactions with scammers also correlated with greater financial losses in the Mainland.
Additionally, the study identified digital media - such as instant messaging apps - as key channels for scam messages. In Hong Kong and Taiwan, the proportion of respondents encountering scam messages on digital platforms was significantly higher than those encountering anti-scam messages (about 50% vs. 40% in Hong Kong; and about 70% vs. about 50% in Taiwan). In the Mainland, exposure rates for scam and anti-scam messages were similar, both at nearly 70%. Notably, among those who responded to suspicious messages, only 35.4% of Hong Kong respondents said they would proactively report suspicious messages or seek help from the police- much lower than the rates in the Mainland (51.4%) and Taiwan (51.2%).
Differences were also observed in the most common types of scams. In Hong Kong and Taiwan, impersonation scams - such as scammers impersonating officials, bankers or relatives - were the most frequently encountered types. However, the scams causing the biggest financial losses were "opportunity-based" scams, such as fake online investment scams. In the Mainland, "consumption-based" and "opportunity-based" scams are the most common types, including never-arrived online purchases and investment scams. "Consumption-based" scams - such as fake online game transactions - lead to the most losses in the Mainland.
Scam persuasion techniques also varied in their impacts. Hong Kong respondents were more susceptible to "scarcity and urgency" tactics (e.g. "limited-time offers") and "commitment" strategies (e.g. "starting with small tasks, then gradually increasing involvement"), making them more likely to respond to scams and experience financial losses. In Taiwan, "scarcity and urgency" tactics similarly had the greatest influence, while in the Mainland, "social influence" (e.g. "everyone is investing in this project") was the most persuasive tactic.
In terms of anti-scam knowledge, Taiwan respondents scored the highest, with an average of 7.1 out of 8, followed by Hong Kong (6.7) and the Mainland (5.1). The questions with the highest error rates were related to everyday financial fraud scenarios, revealing that the public still has significant blind spots when facing scams related to financial security. Given that such scams are both close to daily life and technically complex, people tend to make poor judgments owing to a false sense of familiarity and gaps in professional knowledge - underscoring the urgent need for enhanced financial anti-scam education.
Professor Huang remarked: "Fraud is not simply a matter of poor individual judgment; rather, it represents a long-term battle involving information and trust. Scammers often use prolonged interaction, emotional investment and psychological manipulation to lower victims' defences. Anti-scam agencies must continuously enhance their prevention and response mechanisms. Our findings also show that victims are not necessarily those traditionally perceived as 'vulnerable'; they can also be highly educated individuals that are often considered hard to deceive. This highlights the need for anti-scam education for all."
"Our study found that the longer victims interact with scammers, the greater their financial losses. In both Hong Kong and Taiwan, we also identified a 'critical point effect' between the third and sixth month of interaction." Professor Huang added. "This shows that scam prevention is essentially a race against time. People should end any suspicious contact as early as possible, and immediately reach out to the police, banks, or anti-scam hotlines."
Professor Huang further emphasised: "Strengthening anti-scam knowledge is crucial. Public awareness campaigns must go beyond warnings and advocacy. They should align more closely with everyday life scenarios, and provide easy-to-use detection methods along with actionable guidance to enhance the public's ability to guard against scams. Our research also shows that in areas requiring specialised knowledge, such as financial scams, the public remains particularly vulnerable. This explains why, even with widespread publicity, highly sophisticated and realistic scam tactics can still lead to misjudgement and victimisation."
She stressed: "Building an 'anti-scam ecosystem' that involves the government, media, platforms, businesses and civil organisations is crucial. By reducing scam message exposure and cutting losses in real time across multiple channels, we can significantly strengthen society's overall defence capacity against fraud."
To promote anti-scam education and cross-sector collaboration, CityUHK's Department of Media and Communication has partnered with multiple organisations this year to launch the "Seminar Series on Anti-Scam Research" and actively participates in international anti-scam research projects. In addition, the University's collaboration with the telecom operator on the "Technology Leading, Connecting the World" project has successfully applied big data and AI to enhance scam monitoring and early-warning capabilities. The project was also recognised as an outstanding case at the 2024 World Internet Conference Wuzhen Summit for building a community with a shared future in cyberspace.
CityUHK has also been working closely with the police and telecom operators, forming a cross-departmental task force to promote campus anti-scam education and technological applications. These efforts demonstrate the determination of academia, law enforcement and industry to work hand in hand to combat fraud. CityUHK will continue to leverage innovative technology and social engagement to raise public awareness of scam tactics and dangers, and safeguard digital security together.
The research team also comprised members from the Department of Media and Communication, including Postdoctoral Fellow Muhammad Masood, as well as PhD students Cai Qinxian, Zhang Leyi, Chang Bo, Liu Haodong, Liu Ruoheng, Gao Shuang, and Lyu Xiangxin, along with research assistants Zhan Qinhui and Liang Jinqian.
