top of page

Apr 13, 2026

“Unrivaled Audiovisual AI… Detects All Signs of Danger in Factories” [K-Hidden Champion]

News│Media

■ K-Hidden Champion – LOAS



Equipped with Multi-Channel Sensors and Optical Cameras

Proprietary Robots Collect Factory Noise Data

Detects Subtle Signs Such as Abnormal Temperatures

A Solution Provider for Complex Industrial Safety Inspections


Supplied to Samsung Electronics, LG Electronics, and Hyundai Motor Company

European AI Research Institutes Also Extending Collaboration Proposals


On the 7th, Jaehyun Lee, CEO of LOAS, is seated inside the “AI Square (Acoustic Inspection Software Engine) Laboratory” in Songpa-gu, Seoul, with the industrial inspection robots “Tfoi” positioned on both sides. Photo by Park Yoon-seul / Source: Munhwa Ilbo
On the 7th, Jaehyun Lee, CEO of LOAS, is seated inside the “AI Square (Acoustic Inspection Software Engine) Laboratory” in Songpa-gu, Seoul, with the industrial inspection robots “Tfoi” positioned on both sides. Photo by Park Yoon-seul / Source: Munhwa Ilbo

“We can even detect the sound of a strand of hair caught in an air purifier fan in the middle of Gangnam-daero.”


During an interview held on the 7th at the headquarters of LOAS, an audiovisual AI technology startup located in Songpa-gu, Seoul, Jaehyun Lee stated, “Our company possesses unrivaled technology for controlling acoustics within three-dimensional spaces.”


Founded in 2020, LOAS develops proprietary solutions that utilize robots and drones based on audiovisual AI technology to diagnose abnormalities at industrial sites, supplying these solutions to major Korean corporations.


LOAS’s core solution is the unmanned facility diagnostic integrated monitoring platform “ARQOS.” ARQOS is designed based on the acoustic inspection software engine “AI Square.” Proprietary robots and drones developed by LOAS, equipped with multi-channel microphone sensors and optical cameras, patrol factory environments collecting acoustic information and notifying operators of accidents caused by equipment malfunctions.


The ARQOS monitoring platform (web-based) can rapidly detect abnormal signs at specific work sites. For example, when factory operating sounds differ from normal conditions, the system immediately generates alerts such as “Location Check Required” or “On-Site Action Needed.”


The robots operating in the field are capable not only of acoustic detection but also visual scanning. This allows them to instantly identify various abnormal situations that may occur on-site, including “Abnormal Temperature Detected,” “Intruder Detected,” “Safety Goggles Not Worn,” and “Worker Collapse.”


Source: Munhwa Ilbo
Source: Munhwa Ilbo

Conventional industrial acoustic inspections have traditionally relied on specially constructed anechoic chambers designed to eliminate sound reflections. Jaehyun Lee explained, “When using anechoic chambers, full-scale inspections are difficult, the costs are extremely high, and in the case of industrial facilities, the equipment is often so large that constructing an anechoic chamber is not even feasible.”


The key focus of LOAS is developing AI-powered detection technology capable of operating without such anechoic environments. Lee stated, “If surrounding noise can be patterned to a certain extent through AI technology, inspections can be conducted directly on-site,” adding, “In hazardous or aging industrial sites where workers have difficulty physically accessing the location, LOAS’s solution can become an excellent alternative.”


In practice, LOAS solutions are already being actively utilized by organizations including domestic and overseas affiliates of LG Electronics (China, Thailand, India, and Saudi Arabia), Samsung Electronics, EcoPro, Hyundai Motor Company, Korea Western Power, and Korea East-West Power.


Lee explained, “One major client has numerous chemical gas pipelines installed on the rooftop of its factory, but since the factory was built in the 1980s, the facility had never undergone a proper inspection.” He added, “According to the client’s internal evaluation, if workers were to directly climb scaffolding pipes and manually conduct inspections, the estimated cost would reach approximately KRW 50 billion, making it an extremely expensive area to manage.”


According to the company, when drones equipped with LOAS solutions are utilized instead, the inspection cost is reduced to approximately KRW 180 million, making the technology highly attractive to corporations.


LOAS has also been contacted by and is currently collaborating with Fraunhofer IDMT, the acoustics and media AI research institute under Europe’s world-renowned Fraunhofer-Gesellschaft.


Lee stated, “We are currently working on commercializing AI data possessed by Fraunhofer,” adding, “Fraunhofer visited our company in March last year to conduct technical verification, and we officially began collaboration in November of the same year.”


Approximately 60% of LOAS’s total workforce of 30 employees is dedicated to research and development (R&D). Among the six new employees joining this month, four are assigned to R&D roles.


This year is expected to become a critical turning point as the company anticipates surpassing its break-even point. LOAS is targeting an operating profit margin exceeding 28% and is preparing to secure KRW 10 billion in Series A investment funding this coming June.




Source: Munhwa Ilbo > Reporter Lee Ye-rin


View Full Article>>https://www.munhwa.com/article/11581699




LOAS Inc.

(주)로아스

Contact

Tel. 02-6486-6411

Email. info@loas.ai



​Recommended content

Update

LOAS, a startup innovating industrial sites with acoustic AI that detects abnormal noise in just 0.8 seconds

Apr 7, 2026

MEDIA

Korea Gas Safety Corporation Launches 'Future Technology Strategy Committee'... Establishing Advanced Safety Network Including AI and Robotics

Mar 20, 2026

MEDIA

"Managing Semiconductor Production Hubs from the Sky"… Pyeongtaek City Establishes Drone and AI Safety Management

Mar 17, 2026

MEDIA

[Botgyu Goes] AW 2026 ② | The End of the Era of Standalone Robots? Delivering Insights on ‘Autonomous Manufacturing’ Integrated into Platforms

Mar 12, 2026

bottom of page