Adobe 'Firefly' User Experience
Creating Images for Article Writing Tasks
From Text to Image... Intellectual Property 'Secure'

After erasing part of the photo taken at the cake shop and typing the word 'wonsungi' (monkey), a seamless image was created.

After erasing part of the photo taken at the cake shop and typing the word 'wonsungi' (monkey), a seamless image was created.

View original image

"High-quality images are created in just 5 seconds."


I tried Adobe's generative artificial intelligence (AI) model, Firefly. Adobe is a graphic specialist company famous for Photoshop. So, I thought Firefly might be a work tool only for professional designers, but after actually using it for work, it was enough to break that prejudice.


Recently, I wrote an article to capture the voices of internet companies requesting permission to sell glasses and contact lenses online. I wanted to publish an online article, but it was difficult to find photos that matched the content of the article. There was no image that could simultaneously show 'glasses' and 'online sales.' If I requested the company’s graphic designer, it would take at least 30 minutes.


At that moment, I thought of Firefly. After simply signing up on the website, I entered the 'Text to Image' menu. I wrote in the prompt box, 'Create an image of selling glasses online,' and it generated it in 5 seconds. It was an image of a laptop monitor on a desk with a model holding glasses. I attached the image as a thumbnail to the article, and one reader commented, 'At first glance, I thought it was a photo.' Perhaps because it is a generative AI feature from a graphic specialist company, a clearer and more emotional high-resolution image was created than expected. However, upon closer inspection, the outline of the laptop corner was warped, and the part that looked like the glasses frame was awkwardly expressed. The limitations of AI appeared in the detailed parts.

I used Firefly to find images for an online article. When I typed the desired style in text, a high-resolution, emotional image was generated. However, upon closer inspection, the shape of the glasses is unclear.

I used Firefly to find images for an online article. When I typed the desired style in text, a high-resolution, emotional image was generated. However, upon closer inspection, the shape of the glasses is unclear.

View original image

The free images generated by Firefly have a watermark and can be used commercially. Attaching images or photos that catch the reader’s eye has become an essential element when writing online articles. Firefly seemed sufficient to meet these needs. If the desired result did not come out, it is said that writing prompts △ more specifically △ in a descriptive form △ with emotion helps. It supports more than 100 prompt languages, including Korean.


There is also a feature to recreate images by referencing photos I took. When I uploaded a photo I took at a cake shop, a fantastic image that looked like a cake country was created. After erasing part of the original photo, I filled it with monkeys. A small and cute monkey was sitting on the cake. How about using playful AI features for April Fool’s jokes now?



Adobe explained that Firefly generates new images only based on portfolios and images authorized for use provided by the company. It generates images within the scope that does not infringe intellectual property. Free members can generate 25 images per month. Paid members (Premium Plan, monthly fee 6,600 KRW) can increase it to 100. Adobe announced, "Functions for editing video and audio will also be released soon."


This content was produced with the assistance of AI translation services.

© The Asia Business Daily(www.asiae.co.kr). All rights reserved.

Today’s Briefing