AI Takes Over Public Web Management, Predicting User Complaints in Advance
AI Takes Over Public Web Management, Predicting User Complaints in Advance
Posted January. 08, 2025 11:37,
Updated January. 08, 2025 11:37
While the age of machines replacing human tasks is often discussed, the reality can feel intangible. Despite the integration of AI and automation technologies in various products and services, their work often remains invisible to the human eye.
(Source: Shutterstock)
However, businesses and institutions that have adopted Robotic Process Automation (RPA) solutions—designed to automate repetitive tasks traditionally performed by humans—strongly disagree. RPA solutions, programmed with pre-set scripts, can perform actions like text entry, app execution, and data copying by directly manipulating a keyboard and mouse as if a person were doing it.
Delegating repetitive tasks to RPA can significantly boost efficiency. Examples include security tasks like periodic password updates, web service monitoring involving repeated page refreshes, and alert systems that detect issues and automatically send emails or SMS notifications. These tasks, while simple, are crucial as they require consistent monitoring and response.
Such automation solutions are no longer exclusive to private enterprises known for their early adoption of new technologies. They are increasingly being implemented in public institutions, which are recognizing the importance of minimizing service disruptions and responding quickly to citizen needs.
After last year's government service platform "Gov24" outage, the government has placed a higher priority on preventing and responding to such incidents. Critical government websites are now classified into tiers, with top-tier sites requiring responses to outages within 120–150 minutes.
One top-tier government website adopted an AI-powered RPA solution for web monitoring this summer. This system can quickly detect issues and alert administrators, creating an automated response process.
The solution was developed and supplied by Infofla, a South Korea-based SME specializing in AI-driven IT automation solutions. Founded in 2019, Infofla has gained recognition for its real-time object recognition RPA system, "RPACA (Alpaca)," and its AI-based integrated management platform, "ITOMS." The company has also obtained the GS (Good Software) certification, a critical credential for public sector procurement.
Configuration of the web outage monitoring automation feature using VLAgent (Source: Infofla)
Infofla has tackled the limitations of traditional RPA systems—such as their dependency on rigid scripts that fail to adapt to unexpected variables—by introducing its Vision Language Model (VLM). This model combines large language models (LLMs) with image recognition capabilities.
Leveraging VLM, Infofla recently unveiled "VLAgent," an agent model that perceives screen inputs and executes commands accordingly. This next-generation automation solution merges the latest automation trends into practical applications. Additionally, Infofla launched a free trial of its document inspection and personal task automation services through its web platform (vlagent.infofla.com).
A web service to experience the capabilities of VLAgent (Source: Infofla)
An Infofla representative highlighted the advantages of their solution in an interview: “Gov24 is a highly complex platform used by over 5,000 institutions, making centralized management extremely challenging. AI-based automated monitoring and management are particularly effective here. Our VLAgent solution not only automates repetitive tasks but also plays the role of an 'AI customer,' identifying potential issues before users even notice them.”
With its innovative approach, Infofla is setting a new standard for automation in both public and private sectors. From reducing disruptions to enhancing user satisfaction, VLAgent promises to redefine how organizations handle repetitive tasks and respond to challenges in real time.
BY Kim Yong-Woo (pengo@itdonga.com)
However, businesses and institutions that have adopted Robotic Process Automation (RPA) solutions—designed to automate repetitive tasks traditionally performed by humans—strongly disagree. RPA solutions, programmed with pre-set scripts, can perform actions like text entry, app execution, and data copying by directly manipulating a keyboard and mouse as if a person were doing it.
Delegating repetitive tasks to RPA can significantly boost efficiency. Examples include security tasks like periodic password updates, web service monitoring involving repeated page refreshes, and alert systems that detect issues and automatically send emails or SMS notifications. These tasks, while simple, are crucial as they require consistent monitoring and response.
Such automation solutions are no longer exclusive to private enterprises known for their early adoption of new technologies. They are increasingly being implemented in public institutions, which are recognizing the importance of minimizing service disruptions and responding quickly to citizen needs.
After last year's government service platform "Gov24" outage, the government has placed a higher priority on preventing and responding to such incidents. Critical government websites are now classified into tiers, with top-tier sites requiring responses to outages within 120–150 minutes.
One top-tier government website adopted an AI-powered RPA solution for web monitoring this summer. This system can quickly detect issues and alert administrators, creating an automated response process.
The solution was developed and supplied by Infofla, a South Korea-based SME specializing in AI-driven IT automation solutions. Founded in 2019, Infofla has gained recognition for its real-time object recognition RPA system, "RPACA (Alpaca)," and its AI-based integrated management platform, "ITOMS." The company has also obtained the GS (Good Software) certification, a critical credential for public sector procurement.
Infofla has tackled the limitations of traditional RPA systems—such as their dependency on rigid scripts that fail to adapt to unexpected variables—by introducing its Vision Language Model (VLM). This model combines large language models (LLMs) with image recognition capabilities.
Leveraging VLM, Infofla recently unveiled "VLAgent," an agent model that perceives screen inputs and executes commands accordingly. This next-generation automation solution merges the latest automation trends into practical applications. Additionally, Infofla launched a free trial of its document inspection and personal task automation services through its web platform (vlagent.infofla.com).
An Infofla representative highlighted the advantages of their solution in an interview: “Gov24 is a highly complex platform used by over 5,000 institutions, making centralized management extremely challenging. AI-based automated monitoring and management are particularly effective here. Our VLAgent solution not only automates repetitive tasks but also plays the role of an 'AI customer,' identifying potential issues before users even notice them.”
With its innovative approach, Infofla is setting a new standard for automation in both public and private sectors. From reducing disruptions to enhancing user satisfaction, VLAgent promises to redefine how organizations handle repetitive tasks and respond to challenges in real time.
BY Kim Yong-Woo (pengo@itdonga.com)
About Dong-A Ilbo