"No Need to Hire a Model"... The 25-Year-Old Woman Who Entered Social Media with a Face That Cost Just 1,500 Won [AI Content: The Bright and Dark Sides] 1
A Changing Content Market
Reels Made Without Models or Filming Locations
Production Is Easier, but Trust and Creator Compensation Remain Challenges
"I want to create a Korean woman influencer in her 20s who likes wearing outfits with a Slackercore feel. Draw her face and full-body look together on one sheet."
“Hacienda,” a dark-brown-haired Korean influencer generated by a reporter using AI. Reporter Lee Eunseo.
View original imageThree minutes: that's how long it took this reporter to create "Hacienda," a 25-year-old Korean woman influencer with long, dark brown hair and natural makeup. In ChatGPT Work, the reporter wrote a prompt describing the AI influencer's appearance and clothing. Asked for a "Slackercore" style, with comfortable clothes worn casually, the AI generated images of her face and full body.
The reporter then connected the AI content creation platform Higgsfield as a plugin and attached a draft of its AI influencer feature. To give the character long limbs, a tall frame, and a distinctive look, the reporter chose heterochromia, with a different eye color in each eye. Creating a character who was tall like a model and had gray and brown eyes cost just 1.125 tokens, or 1,500 won.
After attaching a reference video alongside the character and entering, "Make a video of her posing on a vintage street in Euljiro," the AI generated a 1080p Reels video. There was no need to hire a model or find a filming location. It took 10 minutes to generate one video and used 88 tokens, costing 9,500 won. It took just over an hour in total to create Hacienda's Instagram account and upload two Reels videos and a profile photo. The account surpassed 5,000 views just one day after it was created.
However, just as the production process was simple, so too was the process of checking copyright ownership and whether AI had been used. For the most part, it relied heavily on users' own declarations. When photos and videos were uploaded to Higgsfield, a pop-up appeared stating, "The photos and videos used to create content must be owned by you, and you must have permission to use the faces shown." Users only had to acknowledge the clause; there was no separate process to prove ownership of the original material or consent from people pictured. On Instagram, too, users had to indicate for themselves whether they were AI creators and whether their content was AI-generated.
AI Transforms Content Production, Distribution, and Consumption
The social media landscape is changing with the emergence of faceless AI influencers. According to the "2026 AI Content Creation Report," released last August by U.S. digital marketing firm First Page Sage, the share of posts and captions classified as containing AI-generated text has risen rapidly over the past five years. The share grew from 7% to 54% on LinkedIn, from 5% to 41% on X (formerly Twitter), and from 6% to 38% on Facebook. Over the same period, the share of AI posts rose from 4% to 29% on Instagram and from 2% to 23% on TikTok.
Monetization systems that encourage AI content creation are also becoming more sophisticated, making it inevitable that the number of AI influencers will grow. U.S. short-form content platform Launchpoint runs a program in partnership with various AI creation platforms that rewards users for posting AI videos on social media, with payments based on view counts. The program pays $1 per 1,000 views, or approximately 1,300 won, up to a maximum of $2,000, or approximately 2.67 million won, per video.
The Instagram account of overseas AI influencer Jin Phil, for example, is gaining recognition for short videos of the character dancing at Paris Fashion Week, on rooftops, and on the street. The account now has more than 240,000 Instagram followers, and even a meme coin linked to the character has appeared.
AI is boosting productivity throughout the content industry, from production to distribution and consumption. People are still needed to handle planning and editing, but it is now possible to create a variety of locations and scenarios without separate filming equipment or performers. Individuals and small businesses can also produce content at low cost and increase productivity by changing a character's clothes and background to create follow-up videos.
However, the lower barriers to production have also brought significant problems. One prominent example is "AI slop," low-quality content mass-produced with AI, given that platforms provide monetary rewards based on view counts. There is also the risk of deepfakes being misused to create unauthorized composites of real people's faces and voices.
Hot Picks Today
"Who would've thought Gangnam would fall behind?"... The neighborhood drawing buyers in their 30s and 40s with cash [Real Estate A to Z]
- "KOSPI Could Fall to 5,000... Foreign Selling Must Ease" [Weekend Money]
- [Monetary Policy Board Poll] ② "South Korea's Growth to Top 3.5% This Year, Remain in 3% Range Next Year"... 'Inflation' Biggest Variable in Rate Decision
- "I Pay for Comfort"... China's Gen Z Secretly Spend Hundreds of Thousands of Won to Unburden Themselves [China.zip]
- "This Country Has Lost Its Way"... Rapper Nucksal's Criticism of Childbirth Policies Sparks Debate
Moon Hyungnam, a professor in the Division of Convergence and International Studies at Sookmyung Women's University and president of the Korea AI Education Association, said, "Content production will shift toward a collaborative model in which people handle planning and creative judgment while AI supports production and editing, boosting productivity and global reach." He emphasized that "we need to foster an industry ecosystem that enhances both the reliability of AI-generated content and compensation for creators."
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.