Kakao Mobility has set out plans to develop Stage 4 autonomous driving applied sciences in-house as a part of its physical AI technique.
Kim Jin-kyu, vice chairman and head of Kakao Mobility’s Bodily AI division, offered the roadmap on the 2026 World IT Present convention at COEX in Seoul. His session centered on autonomous driving companies constructed round mobility platforms within the bodily AI period.
The occasion was held beneath the title “Past Thought, Into Motion: AI strikes Actuality,” with 460 corporations and organisations from 17 nations collaborating, based on Yonhap. South Korea’s Ministry of Science and ICT additionally described the occasion as linked to a wider bodily AI transition, the place AI is utilized to bodily industrial fields.
Kim stated Kakao Mobility is working to mix autonomous driving applied sciences with bodily infrastructure as a part of its mobility technique in Korea, and goals to ascertain an open autonomous driving ecosystem to assist native competitiveness.
Stage 4 autonomy refers to techniques that may deal with driving in restricted service areas with out requiring passengers to observe the street or take management, based on the US Nationwide Freeway Visitors Security Administration. Such techniques are sometimes deployed in outlined service areas, like autonomous taxi zones or mounted districts.
Stage 4 roadmap
Kakao Mobility’s Stage 4 roadmap is constructed round three know-how areas: machine studying fashions, automobile redundancy, and validation techniques.
The corporate is creating machine studying fashions designed to deal with notion, decision-making, and management with out human enter. These capabilities cowl how an autonomous automobile reads its environment, makes driving choices, and controls motion.
Kakao Mobility additionally plans to make use of automobile architectures with redundant techniques, permitting core capabilities to proceed working if an necessary part fails.
Its validation platform will mix digital simulations with real-world driving information. The system is meant to assist testing, efficiency enchancment, and high quality checks as the corporate develops autonomous driving companies.
Security and management techniques
Kakao Mobility can also be constructing an built-in security administration platform for autonomous automobiles. One part is the Autonomous Automobile Visualizer, a 3D visualisation instrument that shares a automobile’s area of view in actual time and permits passengers to observe driving situations.
The instrument is designed to indicate what the automobile is detecting throughout operation. It reveals passengers the automobile’s driving context throughout a journey.
The corporate plans so as to add a 24-hour management centre and an anomaly detection system utilizing vision-language fashions. These techniques are meant to assist real-time context evaluation, distant intervention, and emergency response.
The deliberate management centre would monitor autonomous driving companies after deployment. Kakao Mobility stated the anomaly detection system will use vision-language fashions, but it surely didn’t present particulars on mannequin structure or efficiency.
Open ecosystem plan
Kakao Mobility additionally outlined plans to share chosen know-how belongings with corporations, startups and producers engaged on autonomous driving.
The belongings embrace large-scale autonomous driving datasets, high-definition (HD) maps, and platform APIs for ride-hailing and dispatch. HD maps assist autonomous driving by offering detailed street data used for localisation and driving choices.
The corporate stated the asset-sharing plan would permit different business members to develop autonomous driving applied sciences with out constructing all of the underlying infrastructure independently.
Kakao Mobility additionally plans to share operational assets, together with fleet administration techniques and on-site response talents. These are a part of the corporate’s plan to assist an open home autonomous driving ecosystem.
Gangnam service information
The corporate pointed to its late-night autonomous automobile service in Seoul’s Gangnam district as one instance of its present work. The service is offered by means of the Kakao T platform, the place customers can entry autonomous driving companies with present mobility choices.
The Gangnam late-night autonomous taxi service recorded 7,754 rides from its launch on September 26, 2024, to February 28, 2026, based on the Seoul Metropolitan Authorities. Town stated no accidents have been attributed to autonomous driving know-how throughout that interval, and the service averaged about 24 journeys per working day.
The service moved from a free pilot to paid operation in April 2026. Seoul additionally expanded the fleet from three automobiles to seven, excluding two reserve automobiles.
The service may be known as by means of Kakao T utilizing both the Seoul Autonomous Automobile icon or the common taxi-hailing menu. Kakao T teams a number of mobility companies in a single app, together with taxi, navigation and vehicle-related companies.
The Gangnam service is accessed by means of Kakao Mobility’s present mobility platform.
(Picture by Hyundai Motor Group)
See additionally: Hyundai expands into robotics and bodily AI techniques
Need to be taught extra about AI and large information from business leaders? Try AI & Big Data Expo happening in Amsterdam, California, and London. This complete occasion is a part of TechEx and co-located with different main know-how occasions. Click on here for extra data.
AI Information is powered by TechForge Media. Discover different upcoming enterprise know-how occasions and webinars here.
