KT Introduces AI to Design and Customer Notices, Cutting Writing and Review Time to 5 Minutes
IT DAILY ·
✦ AI Summary
KT said on the 12th that it has revamped the way it develops digital services such as apps, websites, and text messages.
KT introduced KDS (KT UX Design System) 2.0 and AI agents to standardize screen layouts, notice writing and review, and the application of design criteria.
At Token Factory, the company handled 424 functional requirements, 25 components, and about 1.1 million lines of code, and the actual implementation period was 2.4 months, more than 5 months shorter than the previously estimated 8 months.
BizCrushWant to follow Korean talks and live streams in real time?Try CapCrush on the web
KT said on the 12th that it has revamped the way it develops digital services such as apps, websites, and text messages. KT unified, under a single standard, the screen layouts and notice copy that had varied by service across its digital services, using AI as part of the revamp. The purpose of the overhaul is to improve customer convenience and ease of use when accessing KT services.
As part of the process, KT built the UX design system KDS(KT UX Design System) 2.0. KDS(KT UX Design System) 2.0 is a system that strengthens AI-enabled functions, and KT employees use it to design digital service screens. KT applied AI to design work to increase consistency in service screen design.
KT also introduced an AI agent to support the drafting and review of customer notice messages. KT employees use the AI-based "message generation agent" to review customer notice messages. KT is also applying AI to customer communication work to increase consistency in notice copy and push ahead with improving customer convenience and ease of use in KT services.
KDS is the common standard for screen layouts, design elements, and notice copy in KT digital services. KDS 1.0 was released internally last year, and the design principles were established based on customer research and other inputs.
A key feature of KDS 2.0 is that AI can directly use the standards. KT cited as the biggest change in KDS 2.0 that AI, too, can now read and use design guides that had previously been viewed only by people.
To do this, KT built the "KDS Harness." The "KDS Harness" ensures that AI composes screens that conform to design standards and also checks AI outputs.
KT said that this system reduces repetitive design work in the service development process and eases the burden of revisions. It also produces the effect of applying consistent screen layouts and a consistent user experience across various services.
KDS is already being applied to real services. Examples include My KT, KT.com, and Direct Shop, which were revamped in September this year, and it is also being used in the AI model platform "Token Factory."
Using KDS, Token Factory increased development efficiency by running planning, design, and development in parallel and automating repetitive coding and documentation. With this approach, it handled 424 functional requirements, 25 components, and about 1.1 million lines of code.
The work was estimated to take about 8 months under the existing method, but the actual implementation period was 2.4 months. As a result, the development period was shortened by more than 5 months.
Meanwhile, KT is improving how it delivers customer information in addition to screen design for digital services. KT sends about 170 million text messages each month, and those messages contain information on fees, service use, and various benefits.
To improve comprehension of the large volume of notice messages it sends, KT has updated its UX writing guide, which sets standards for customer notice copy. It is also replacing telecom jargon with easier expressions; as an example of the terminology changes, it has changed "prepayment" to "pay in advance."
KT officially launched its in-house AI-based "message generation agent" on October 30. The company said the rollout was intended to put the standards into actual work. The "message generation agent" can draft and review customer outbound messages so that they comply with KT writing standards and relevant legal standards.
The tool supports drafting, review, images, and translation in one workflow. It supports RCS messages and recommends images suitable for the content. RCS messages can include images and buttons, and multilingual automatic translation is also supported. Previously, each person in charge handled writing and review, image production, and translation separately, but after the change, entering the key content prompts AI to support the related tasks.
As a result, the time required has also been greatly reduced. The time needed for message drafting and review has been cut from 30 to 60 minutes to about 5 minutes. The time needed to produce images has also fallen from about 4 hours to 15 minutes.
With KDS 2.0 and AI agents, KT is shifting the entire design and customer communication workflow to an AX approach. AI support is limited to repetitive tasks and standard checks. As a result, employees can focus on creative and higher-value work such as analyzing customer needs and improving service quality.
The company plans to further advance AI-enabled design standards and also further improve its review system for AI-generated outputs. At the same time, the company plans to expand KDS adoption to external partners, to various digital channels, and to AI devices as well.
Source: IT DAILY · Seong Won-young
Original: https://www.itdaily.kr/news/articleView.html?idxno=242864
References
This article was produced with the help of an automated content generation algorithm.
Source: IT DAILY
View originalThis article was summarized and organized by BizCrush based on the original article from IT DAILY. For exact quotations and full details, please refer to the original article.
Keep reading
BizCrush