In November 2024, a mother and daughter in Taipei fell victim to a scam after encountering an advertisement on Facebook featuring a fake celebrity. They were swindled out of NT$12 million (US$0.37 million), and following apparently dismissive treatment by the police, the stress became unbearable, leading them to tragically take their own lives.

This incident highlighted a growing issue: Why are scams proliferating, and why do social media platforms remain fertile ground for fake celebrity endorsements and ads promoting scams or fake services?

Facebook: The Scam Exposure Leader 2 Years Running

At the root of the problem is the flood of fraudulent advertisements on social media platforms.

According to the latest survey by the Global Anti-Scam Alliance (GASA), 55 percent of Taiwanese respondents said they encountered scam messages in 2023 through social media postings, exceeding the global average of 44% and making social media the fastest growing source of scams in the local market.

It also meant that social media overtook text messages as the second most common source of scams in Taiwan, trailing only phone call scams (encountered by 60 percent of respondents).   

The survey also found that Facebook stood out as the leading social media platform for scams for a second straight year, with more than 60 percent of Taiwanese respondents having encountered fraudulent messages on the Meta platform.

"Meta is our biggest headache right now," said Digital Affairs Minister Huang Yen-nun (黃彥男) in an interview with CommonWealth Magazine.

Huang found that of the average daily losses of NT$400 million (US$12.2 million) to scams posted by the Criminal Investigation Bureau, nearly 70 percent stemmed from Facebook ads, consisting primarily of investment scams and product endorsements  by fake celebrities.

Fake ads reported to the police put an even greater onus on Facebook, with 97.9 percent originating from the Meta platform and only 2.1 percent from Google. 

Yet, when Huang has requested that Meta remove the ads for scams, the company has often taken two to three days to respond.

Meta has attributed the delays to its technical team being based in the U.S., meaning that when scams are reported during the day, it is the middle of the night in America.

"Fraudulent ads spike on weekends because the U.S. team doesn’t work then," Huang said wryly. He also suggested that Meta’s ad-revenue model — where longer ad exposure means more profit — might be contributing to its lukewarm efforts to combat scams.

The “Cultivate, Trap, Kill” Model: LINE Caught in the Crossfire

One of the most prevalent online scams in Taiwan involves enticing users to click on fake celebrity ads on Facebook. Victims are then redirected to LINE groups, where scammers employ a “cultivate, trap, kill” strategy to solicit money for fake investment opportunities.

"LINE is also a victim in this industrial chain and doesn’t earn a cent from it," said Jofan Lin (林若凡) , head of corporate affairs at LINE Taiwan Limited. 

Unlike Facebook, LINE requires that ads be reviewed by people before being posted, preventing fraudulent ads from appearing on its platform.LINE still struggles, however, with a proliferation of scam accounts. In March 2023, LINE’s cybersecurity team discovered that many users had added fake accounts promoted via Meta and YouTube ads.

In response, LINE developed a crawler system to scan for fraudulent ads on other platforms and identify linked LINE accounts, which are then manually deactivated. As of October 2024, LINE had deactivated 11,451 scam accounts.

"If other platforms had implemented stricter controls, we wouldn’t have to work so hard," Lin admitted.

Taiwan’s Ministry of Digital Affairs currently employs a multidimensional approach in combating scam ads. It encourages public reporting via the "Internet Fraud Reporting and Inquiry Network” app and also uses AI programs to actively scan and identify scams. Once the ads are confirmed as scams with the police, the Financial Supervisory Commission, or impersonated celebrities, platforms are required to take them down.

In May, when the MODA app was launched, there were more than 80,000 scam ads reported per week. By the first week of December, the number had dropped to just 7,000.

The amount of financial losses reported to the anti-fraud hotline has not significantly decreased, however, remaining between NT$11 billion (US$33.6 million) and NT$13.8 billion  (US$42.3 million) a month.

"It takes only five seconds for a scam group to create a fraudulent ad. The numbers will never reach zero," Huang conceded. Scammers also adapt quickly; when investment ads are regulated, they switch to fake celebrity book giveaways. To truly strengthen prevention measures, a third strategy is required: mandatory identity verification for ads.

Yen-Nun Huang, the minister of Ministry of Digital Affairs. (Photo: Pei-Yin Hsieh)

Ad Identity Registration to Begin in January 2025

Under Taiwan’s Anti-Fraud Crime Prevention Act, passed in July, online platforms will be required to disclose the identity of the entity that placed or funded an ad for every ad posted or promoted. The information to be disclosed includes the name of the individual or the name of the corporate entity and its address. 

The new act sets stiff penalties for violations. Digital platforms must establish Taiwan representatives, remove fraudulent ads within 24 hours, and suspend offending accounts, or face a fine of up to NT$100 million (US$3.06 million).

Globally, only the U.K. and Australia mandate identity registration for financial and investment ads, while the EU requires it for political ads. Taiwan is the first to extend identity requirements to all ads. Enforcement begins in 2025, and will target platforms such as Google, LINE, Meta, and TikTok.

Google and LINE have already instituted identity verification for ads, while Meta has pledged to have its registration system ready by Jan. 30, 2025.

Huang said the purpose of the identity verification system is "traceability." If scammers post fraudulent ads, law enforcement authorities can use electronic signatures or FIDO authentication to trace and deter these criminal groups.

Challenges to Implementing the System

This identity verification system faces two major challenges.

First, enforcement is uncertain. Despite a revision to the Securities Investment Trust and Consulting Act in 2023 requiring identity verification for investment-related ads, fraud persists.

Trusted service provider Gogolook observed that many ads on Facebook and Instagram have used disbanded companies or unverifiable entities as advertisers since the system took effect in August. 

Some ads promoting investment tutorials and carbon-credit stocks, for example, have listed the ad buyers as BIGBAKER LLC or “Fast Company” (費思特有限公司) while using the logo of a real financial services provider, Yuanta Financial Holdings.

BIGBAKER LLC’s website, however, offered no clear business information, and “Fast Company” has disbanded.

Second, the vast majority of scam ads originate overseas. Wang Ji (王濟), the subdivision chief of the Anti-Fraud Center under the Criminal Investigation Bureau, estimated that at least 80 percent of fraudulent ads are foreign.

"It’s obvious these foreign accounts posting financial ads for Taiwan are fake," said an indignant Wang, whose division reviews thousands of scam ads a day. 

Taiwan’s unique international status limits judicial cooperation, however, leaving many overseas scammers beyond the reach of the law.

"If we can’t even apprehend them, how can we punish them?" wondered Hsu Fu-shen,  a professor in Central Police University’s Department of Administration Police. He argued that the MODA was “too optimistic” about the prospects for the identity verification system. 

Huang replied that his ministry would continue to evaluate the system’s effectiveness in blocking fraudulent ads, and if there was no improvement within three to six months, the ministry could impose stricter regulations on foreign advertisers. That might entail requiring foreign advertisers to partner with Taiwanese agents or individuals to be allowed to run ads locally.

Time for Social Media Platforms to Step Up

Ultimately, Meta remains at the root of the problem. Huang said Meta has had the tools necessary for identity verification, such as the blue checkmarks for celebrity accounts, but has not applied them to ads. 

"They exploit legal loopholes to maximize profits. That is unacceptable," he said.

Huang vowed that “I will impose penalties” if people are still being scammed through Facebook ads after January 2025. 

When CommonWealth contacted Meta for comment, it did not respond. Ultimately, social media platforms can do a lot to fight scams; it’s just a question of whether they are willing to do so.

LINE Taiwan, for example, recently developed an automated account deactivation process that automatically reviews reports of fake LINE accounts from police agencies, prosecutors offices, and investigation bureaus. Launched in November 2024, the system deactivated 2,000 fake accounts in one month, with a 100 percent accuracy rate.

LINE’s Lin said the company plans to roll out this system in other countries, including in Japan, Indonesia, and Thailand.

"Taiwan’s scammers are ahead of the curve, and our anti-fraud measures should be too," Lin said.

In fact, Taiwan’s laws are starting to make strides in combating fraudulent ads, but if social media platforms do not shoulder their responsibility in joining the fight, it will be hard to beat the scammers who are ripping off Taiwanese people every day.

Have you read?

• Taiwan Faces Uncertainty Amid Trump’s Trade Tariff Plans
• How Can Labor-starved Taiwan Attract More Foreign Talent?

Edited by Luke Sabatier
Uploaded by Ian Huang