In February of 2024, CNN reported a major case of fraud in Hong Kong. After a video­conference between an HK-based finance worker and his company’s chief financial officer (CFO) at corporate headquarters, nearly HK$200 million (about US$25.6 million) was transferred to the CFO. It was later discovered that the so-called CFO was actually a scammer using “deepfake” techniques.

“Fraudulent activity will only increase, using methods that are continually being updated,” says Marco Tsai, public relations and communication manager at anti-fraud company Gogolook. Scammers prey on the human weaknesses of fear and greed to get people to drop their defenses and transfer funds without first confirming the transaction.

Gogolook’s main product is Whoscall, a mobile app that identifies unknown incoming phone calls and text messages. Thus far this app has been downloaded 100 million times worldwide and its service now extends to nine Asian countries. It is very popular in Taiwan, where one in two mobile device owners is an active user of Whoscall. Since its founding in 2012, Gogolook’s revenues have continually increased, with turnover in 2023 reaching NT$770 million and annual growth of 83%. This indicates that in an era when scams are rampant worldwide, the future looks bright for fraud prevention businesses.

Gogolook cofounder and CEO Jeff Kuo received a Presidential Innovation Award in 2020.

Reputation boost

Whoscall received a huge boost to its reputation when it was mentioned in a speech made in Taiwan by former Google CEO Eric Schmidt. Besides becoming the subject of endless media stories and interviews, its creators also received an injection of funding from venture capital firms. The three writers of the app, Jackie Cheng, Jeff Kuo, and Reiny Song, who had been creating apps part-time, decided in 2012 to quit their jobs and dedicate themselves fully to Whoscall. Seeing that more and more people were looking at their cell phones as they walked, they founded a company that they called “Gogolook.”

Marco Tsai, public relations and communication manager at Gogolook, reveals: “Of Gogolook’s nearly 200 employees, 100 are responsible for back-end maintenance and operation of products and databases. This is not easy to achieve in either the public or the private sector.” (photo by Kent Chuang)

User cooperation, technological alerts

Gogolook thus rose from obscurity to popularity overnight and from bleak prospects to receiving an investment of NT$529 million from Naver (the parent company of the messaging app Line). Confidence in the company comes from the fact that it removes the uncertainty, unease, and information asymmetry of incoming phone calls and messages.

Marco Tsai notes: “Our database is at the core of our technology, and within this core telephone numbers are the main protective ‘moat.’ Gogolook currently has 2.6 billion phone numbers, and in recent years we have begun to extend our services to website domain names. This is the largest fraud prevention database in East Asia.” Information from users, who have become a collaborative community, accounts for some 20–30% of this data.

Gogolook’s main partner in Taiwan is the Criminal Investigation Bureau of the National Police Agency. Tsai mentions fraudulent phone calls received by consumers related to installment payments for online purchases as a case in point. Whoscall will indicate “this is a high-risk number” and remind users to be alert and to verify the identity of the caller before transferring money.

This is why the police tell elderly people who may be vulnerable to being swindled that they should “install Whoscall, because it can help you identify phone calls from scammers.”

Royal Thai Police also onboard

Police agencies in Thailand, Malaysia, the Philippines, and Hong Kong have also recommended that their citizens use Whoscall.

Tsai points out that there was a massive increase in the number of fraud cases in Southeast Asia in 2020 to 2021, and local residents began widely using Whoscall’s services, with an explosion in recommendations.

Gogolook also works with the Royal Thai Police and the Thai National Cyber Security Agency. “The Thai police also recommend Whocall to citizens. For us, these are recommendations that money can’t buy,” says Tsai.

Gogolook’s services are available in nine markets including Japan, Korea, Malaysia, and Brazil. Besides having the highest penetration of any anti-fraud app in Taiwan, Hong Kong, and Thailand, in 2024 the number of users in Thailand is projected to surpass that in Taiwan. In 2023 Gogolook became a founding member of the Global Anti-Scam Alliance, an organization that forms a bridge between the anti-fraud industries of Asia and Europe.

In February 2024 Gogolook began working with StoreFront, a Japanese retail chain with more than 2,000 outlets, to link up with Gogolook’s phone number data­base and create local caller ID services.

Gogolook became the first Taiwanese software company to list on the Taiwan Innovation Board. Company staff marked the occasion with a group photo outside the Taiwan Stock Exchange.

Leader in corporate scam prevention

In 2023 Gogolook was named by the National Development Council as one of Taiwan’s “Next Big” rising-­star startup companies and also became the first software firm in Taiwan to list on the Taiwan Innovation Board of the Taiwan Stock Exchange.

“We develop ‘Trust-Tech’ [trust technology] and we are striving to solve the problem of scamming,” says Marco Tsai. Gogolook, which made its name as a company that combats fraud using information and communications technology, realized that the “last mile” for financial scammers of all kinds involves contacting potential victims by phone call or message. However, while technology can identify the source of this essential factor, it is even more important that every­one should “be alert and cautious” in order to guard against fraud.

In 2023 Gogolook held the 1st Anti-Scam Asia Summit, inviting police agencies and NGOs from various Asian countries to engage in transnational cross-domain dialogue.

Authme

The verb for Merriam-Webster’s 2023 word of the year “authentic” is “to authenticate.” Authme, another startup, got its name from the words “authenticate me” (confirm my identity). Authme’s business model focuses on digital identity verification. Building on AI, it can unmask deepfake technology and authenticate a person’s real identity.

Gogolook is working with Japanese retail chain StoreFront to support the latter in developing their own brand of caller-ID services. The photo shows Gogolook CEO Jeff Kuo (left) with StoreFront CEO Hideaki Okada.

Facial recognition

“It’s easier to understand the importance of digital identity authentication by taking an example from real life,” says Yvonne Lin, Authme’s head of marketing. Right now when opening an account with a financial institution it is popular to apply online, and anyone can do so by providing a photo of their ID document, entering their personal data, and uploading a photo of their face for identity verification. But how can the financial institution be sure that the ID supplied by the applicant is genuine, that the applicant is the same person as is shown on the ID, or that the applicant is even a real person?

“This is the so-called ‘KYC’ [know your customer] problem for the financial industry, and they can face scams in any part of the process,” says Lin. For example, IDs can be misappropriated or faked. Using AI, Authme can detect anti-fraud measures such as laser labels on identity documents and determine whether the ID is real.

For another example, selfies can be faked using AI software like ChatGPT Sora. How can they be verified? Again using AI, Authme first compares the individual’s face to the photo on the ID card to see whether they match, and then uses “passive liveness detection” technology to determine whether or not the person in front of the camera has the attributes of a living person. For instance, is blood flowing through the capillary blood vessels in the skin below the eyes? Is the skin texture correct? The whole process seems very complex, but in fact it can be completed within one minute.

Gogolook has signed a memorandum on cooperation with Thailand’s National Cyber Security Agency.

Opportunities for ethical hackers

Authme is able to find business opportunities for identity authentication at all the points where fraud can be committed because its cofounder and CEO Andy Lee was once a “white hat” (ethical) hacker—he was a software engineer for data leak troubleshooting. In 2016 he founded a trading platform for virtual currencies, and thought from the viewpoint of a hacker about the ways in which it might be vulnerable to false identity attacks. He discovered that there were two major problems with identity verification.

The first was that the process was inconvenient. The second was that businesses used human verification, but as technology had evolved the human eye could not always accurately detect faked photographs.

Lee identified a business opportunity and invited fellow ethical hacker Dalton Hsu, along with Tseng Kuo-chan, a businessman who promotes the ideal of financial inclusion (equal access to financial services for all), to join him, and in 2019 they cofounded Authme to use AI technology for digital identity authentication. Coincidentally, it was just at that time that Taiwan began to authorize Internet-­only banking, and Authme began working with Line Bank, the first Internet-­only bank to be licensed in Taiwan, to conduct digital identity verification. To date this system has completed more than 3 million identity verifications.

Authme CEO Andy Lee thinks about solutions to identity authentication from a hacker’s point of view.

(photo by Kent Chuang)

Overseas markets and financial inclusion

Looking overseas, the main potential competitors in the facial recognition market are European and American firms. Yvonne Lin notes that Caucasians and East Asians have different facial bone structures, and Authme’s competitive advantage lies in its dataset of more than 1 million Northeast-Asian and Southeast-Asian faces. ­Using this dataset, it can train AI models and optimize them through federated learning to achieve 99.7% accuracy.

Besides facial recognition, Authme also performs ID document verification. The company scans the personal information on passports using optical character recognition (OCR) technology, and can then directly read the passport chip using the near-field communication capability of mobile phones. Lin says that Authme has already patented this technology in Europe, America, Japan, and Korea, and hopes to employ it to cross over into overseas markets.

Authme chief operating officer Tseng Kuo-chan previously ran an aquaculture business in Indonesia, and he saw that in many remote places in that country there were no banks, thereby depriving local residents of the opportunity to access financial services and get loans. But since nowadays everyone has a mobile phone, they should be able to use that technology to enjoy the benefits of financial inclusion. Through the optimization of its AI models, Authme can meet the requirements of Southeast-Asian markets with respect to locations with limited Internet bandwidth.

With the increasingly widespread use of digital applications, identity verification remains crucial as the first line of defense to deny scammers the opportunity to victimize people. Andy Lee concludes: “This is a problem facing the whole world, but Taiwan is outstanding in this area. Due to our geopolitical situation we often come under attack, but technology is our hard power, and if we can find solutions to the problem here, we can promote them around the world.”

Using “passive liveness detection” technology, Authme can assess whether the figure in front of a camera is a living person or not. If a screen image is used instead, a warning will appear indicating that this is not a real individual. (photo by Kent Chuang)

Yvonne Lin, head of marketing at Authme, says that their AI technology can quickly determine whether a personal ID document is genuine or not. (photo by Kent Chuang)

Authme specializes in the identification of Asian faces, using technology to verify who a person is. Not only is the process efficient, it can identify fakes that the human eye cannot see. (courtesy of Authme)