Meta violated fraud prevention regulations twice on the same day and was fined NT$10 million in a single day

Meta was fined twice on the same day on August 14 this year for violating the "Regulations on the Prevention of Fraud and Crime Harm." Both cases were fined NT$5 million, for a total fine of NT$10 million in a single day. (Reuters)

[Reporter Qiu Qiaozhen/Taipei Report] Supervision of fraudulent advertising on online platforms continues to tighten, according to the administrative penalty records recently released by the Digital Industry Administration (Digital Industry Administration) of the Ministry of Digital Development (https://web.moda.gov.tw/ADI/antifraud/admin0620/ad/16520 ), the social platform Meta was fined twice on the same day today (August 14, 2026) for violating the "Regulations on the Prevention of Fraud and Crime Harm." The two cases were fined 5 million yuan each, and the total fine was 10 million yuan in a single day. The violations involved the information disclosure of AI, deep fake personal image advertisements, and the obligation to deal with fraudulent advertisements.

One of the sanctions involves Article 31, Item 1, Paragraph 4 of the Regulations on the Prevention and Control of Fraud Crime Harm. According to the provisions, when online advertising platform operators publish or promote advertisements, they should disclose "whether the advertisement uses deep forgery technology or personal images generated by artificial intelligence."

Please read on...

The Ministry of Industry and Information Technology stated that Meta violated the above-mentioned regulations by failing to disclose information related to the use of deep forgery technology or artificial intelligence to generate personal images in its advertisements. Therefore, it imposed a fine of NT$5 million in accordance with Article 40, Paragraph 2 of the Ordinance, and ordered it to make corrections within a time limit.

It is worth noting that Article 40, Paragraph 2, applies to those who violate the provisions of Article 31, Paragraph 1, and the circumstances are serious, they may be fined not less than 1 million yuan but not more than 25 million yuan, and ordered to make corrections within a time limit; those who fail to make corrections within the time limit may be punished on a case-by-case basis. In other words, Meta’s violation has been deemed to be serious by the competent authorities.

Another fine of NT$5 million related to the handling of fraudulent advertisements. Meta violated Article 32, paragraph 1, paragraph 1, of the "Regulations on the Prevention of Fraud and Crime Harm," and was fined NT$5 million in accordance with Article 39, paragraph 1, paragraph 3, of the Regulations.

According to Article 32, Paragraph 1, Paragraph 1, when an online advertising platform operator learns that an advertisement it publishes or promotes is a fraudulent advertisement or is obviously involved in fraud, it shall remove, restrict browsing, stop broadcasting of the advertisement, or take other necessary measures, either on its own initiative or within a period notified by the judicial police agency, the digital economy-related industry competent authority, or the competent authority of various industries, and provide relevant information to the judicial police agency.

However, the Ministry of Industry and Commerce did not further explain how many fraudulent advertisements, overdue dates and related case details were involved in this violation. It only stated that it would not specifically disclose the details of individual cases this year.

Meta has been fined for the handling of fraudulent advertisements in the past. In August 2025, the Ministry of Digital Affairs conducted an investigation into Facebook's handling of fraudulent advertisements. At that time, the Taipei City Government referred 1,763 cases, of which 77 were not removed within 24 hours. The Ministry of Digital Affairs finally fined Meta 2.5 million yuan for violating Article 32, Item 1, Paragraph 1, of the "Fraud Crime Harm Prevention and Control Ordinance".

In addition to the handling of fraudulent advertising, another 5 million yuan fine this time highlights new regulatory issues brought about by AI-generated content. With the rapid development of generative AI technology, the issue of using celebrity images to produce investment, financial management and other advertisements has attracted attention. The "Regulations on the Prevention and Control of Fraud and Crime Harm" has included personal images generated by deep forgery technology or artificial intelligence into the scope of disclosure of online advertising information.

This time, Meta was fined 5 million yuan for violating two different regulations on the same day, for a total fine of 10 million yuan in a single day. According to the public administrative punishment information of the Ministry of Industry and Information Technology, since the "Fraud Crime Harm Prevention Regulations" came into effect, Meta has been fined many times for issues such as the disclosure of advertising information and the handling of fraudulent advertisements, including: two transactions of 500,000 yuan each in May 2025, and an advertisement in June of the same year. The company was fined RMB 15 million for a lack of service management system and the circumstances were serious. In August of the same year, it was fined RMB 2.5 million for failing to dispose of fraudulent advertisements in accordance with the statutory notice period. In addition to two fines of RMB 5 million each on August 14 this year, the total amount of fines has reached RMB 28.5 million.

Meta response: Close cooperation with the Department of Digital Information and law enforcement agencies

In this regard, this newspaper asked Meta, and a spokesperson responded, "Fraud syndicates use various channels to defraud people and constantly change their tactics to evade law enforcement actions. Meta works closely with Taiwan's Ministry of Digital Development and law enforcement agencies to remove fraudulent advertisements. Of the URLs that were notified, more than 99% have been processed within the stipulated 24 hours. For the very few remaining cases, we continue to cooperate with relevant agencies to deal with them."

Regarding AI-generated content, a Meta spokesperson said, "We will add an 'AI Information' label to ads that are produced or heavily edited using Meta's generative AI tools, as well as ads that are produced or edited by third-party AI tools that are detected through industry standard signals. We are committed to cooperating with the Digital Development Department and the Criminal Police Department to jointly combat fraudulent advertising in Taiwan."

As for how to identify and label AI-generated ads, according to Meta's official website information (https://about.fb.com/news/2025/02/gen-ai-transparency-metas-ads-products/amp/?utm_com), the platform will use "content traceability metadata" that meets industry standards to detect whether the ads are generated by third-party AI tools and automatically label them. In addition, it will also use the advertiser's independent disclosure mechanism to identify and label AI-generated images in ads.

Grasp the economic pulse with one hand. Click here to subscribe to Free Finance Youtube channel