02/03/2026

LYFE MONDAY | MAR 2, 2026

23

A NEW trend has gone viral on social media – users share a personal photo and ask artificial intelligence (AI) tools to create a caricature or illustration based on their life, their job and “everything the AI knows” about them. The result showing animated versions of the person at the office, with their family, or representing their profession, has become frequent content on Instagram, TikTok and LinkedIn. The popularity comes at a time when Malaysians are among the most enthusiastic AI users globally. According to the Ipsos AI Monitor 2025 survey, 77% Malaysians say AI products and services make them excited – placing Malaysia third out of 30 countries in excitement over AI usage. While the trend may seem creative and entertaining, cybersecurity specialists warn that this practice can expose personal information and enable the creation of personalised, large-scale fraudulent messages – a very real threat today. According to Kaspersky experts, this type of request does not work like a simple visual filter. To achieve more accurate images, people allow AI tools to access all the information associated with their profiles without restrictions, since the instruction itself is embedded in the command “create a caricature about me and my job based on everything you know about me.” In addition to the reference photo, extra data such as company name, corporate logos, job title, city, daily routines, hobbies and other family details are often included and used to create the trend. Each of these data points is a key piece in building a detailed digital profile. By combining image, text and context, habits, relationships, frequently visited places and professional responsibilities are revealed – information that cybercriminals can exploit to craft more sophisticated scams. As a result, a fraud attempt that mentions where someone works, their job title or even a family member becomes far more convincing and increases the likelihood that the victim will trust it and share sensitive information or money. In addition, when interacting with these platforms, users are not only sharing the final image. Depending GOOGLE unveiled a successor to its viral Nano Banana image generation model, touting faster performance as the search-engine behemoth pushes to attract more users to its AI tools. The model, called Nano Banana 2, is being rolled out across products, including the Gemini app, AI Mode and Lens features on Search, and Flow, its AI-powered video tool, the Alphabet-owned company said. The launch is Google’s latest in a series of moves that have propelled the company to the forefront of the AI race, helping it better compete with OpenAI’s ChatGPT after suffering a string of embarrassments initially. The success has fuelled a 47% surge in its stock in the past six months. Google rolled out the Nano Banana AI image editor in August, which quickly became a viral

AI tools can be told to generate caricatures based on a person’s job. – AI-GENERATED IMAGE FROM CHATGPT

Beware trending AI caricature vanity trap, experts warn

facades or any element that could help locate you or associate you with an organisation. 0 Do not share information or images of minors, nor reveal family details that could be used to impersonate close contacts or design emotional scams. 0 Review the platform’s privacy policy and permissions before using it, especially regarding content retention and the use of data for training or service improvements. 0 Complement caution with active digital protection: Solutions such as Kaspersky Premium help reduce risks from malicious links, dangerous downloads and phishing techniques associated with these trends, while reinforcing the security of the devices used to create and share this content.

o Allowing tools access to personal information, such as job titles, to generate images can facilitate identity fraud, social engineering attacks

even a cautious user’s defences,” said Kaspersky managing director for Asia Pacific Adrian Hia. While these tools can be a fun way to experiment with digital creativity, experts recommend adopting more cautious habits when participating in this type of trend. To reduce risks: 0 Avoid entering identifiable data in prompts, such as full name, job title, company, city, address, schedules or routines – even if it seems “just to personalise” the image. 0 Do not upload photos that show logos, credentials, documents, licence plates, screens, building

voluntary for cybercriminals. Every time users in Apac prompt an AI with details about themselves just to see a clever illustration, they are handing over the blueprints for a perfect social engineering attack. “In a region where AI adoption is leading the world but technical literacy is still catching up, these digital portraits are becoming dangerous maps. We are essentially giving scammers the ‘context’ they need to turn a generic phishing email into a highly convincing, personalised scam that can bypass briefing

on the service and its privacy policies, the original photo, the text or instructions written by the user, usage history and certain technical data – such as IP address, device or interaction patterns – may also be stored. Part of this information may be retained to operate the service, improve performance or train AI models, meaning the content does not necessarily disappear after the caricature is generated and may remain longer than users expect. “This viral trend of caricature creation of our lives may seem like harmless fun, but it is effectively a

Google rolls out Nano Banana 2 after success of image generation tool sensation, attracting 13 million first-time users to the Gemini app in just four days in September. By mid-October, it had generated more than five billion images.

The tech giant followed that with the release of the upgraded Nano Banana Pro in November. Nano Banana 2 leans on Gemini’s faster and cheaper models known as Flash, which allows quicker image generation and editing, Google said, adding that it also has better instruction-following capabilities and delivers sharper details. In November, Google released its Gemini 3 AI model, whose success prompted rival OpenAI to issue an internal “code red” to push teams to accelerate development. Gemini 3 has significantly boosted user engagement, helping the Gemini app capture more than 750 million monthly active users at the end of December. – Reuters

Nano Banana 2 leans on Gemini’s faster and cheaper models known as Flash. – AI-GENERATED IMAGE FROM NANO BANANA 2