Kakao Mobility: Ambitious Strategy for Level 4 Autonomy

Kakao Mobility Unveils Its Vision for the Future of Autonomous Driving

At the World IT Show 2026 held at COEX in Seoul, Kakao Mobility unveiled its ambitious plans for the development of Level 4 autonomous driving technologies. Kim Jin-kyu, Vice President and Head of the company's Physical AI division, presented this roadmap as part of their physical AI strategy. The event, titled "Beyond Idea, Into Action: AI moves Reality," attracted 460 companies and organizations from 17 countries, according to Yonhap.

The South Korean Ministry of Science and ICT emphasized that this event is part of a broader transition towards physical AI, where artificial intelligence is applied to physical industrial sectors. Kim explained that Kakao Mobility is striving to combine autonomous driving technologies with the existing physical infrastructure in Korea, aiming to create an open autonomous driving ecosystem that will enhance local competitiveness.

Understanding Level 4 Autonomy

Level 4 autonomy, as defined by the U.S. National Highway Traffic Safety Administration, refers to systems capable of managing driving in limited service areas without passengers needing to monitor the road or take control. These systems are typically deployed in defined service areas, such as autonomous taxi zones or fixed districts.

Kakao Mobility's roadmap to achieve this level is based on three technological pillars: machine learning models, vehicle redundancy, and validation systems. The company is developing machine learning models capable of handling perception, decision-making, and control without human intervention. These models enable the autonomous vehicle to perceive its environment, make driving decisions, and control its movements.

Safety and Technological Redundancy

Kakao Mobility also plans to utilize vehicle architectures equipped with redundant systems, ensuring that essential functions continue to operate even in the event of a critical component failure. Simultaneously, its validation platform will combine virtual simulations with real driving data to support testing, performance improvement, and quality checks as the company develops its autonomous driving services.

Advanced Safety and Control Systems

To ensure the safety of autonomous vehicles, Kakao Mobility is building an integrated safety management platform. A key element of this platform is the Autonomous Vehicle Visualizer, a 3D visualization tool that shares a vehicle's field of view in real-time, allowing passengers to monitor driving conditions. This tool is designed to show what the vehicle detects during operation, thus informing passengers of the driving context.

The company also plans to add a 24/7 operational control center and an anomaly detection system using vision-language models. These systems aim to support real-time contextual analysis, remote intervention, and emergency response. The control center will monitor autonomous driving services after their deployment, although Kakao Mobility has not yet provided details on the architecture or performance of the anomaly detection models.

Sharing Technological Assets for an Open Ecosystem

As part of its plans for an open autonomous driving ecosystem, Kakao Mobility announced its intention to share some of its technological assets with other companies, startups, and manufacturers working on autonomous driving. These assets include large-scale autonomous driving datasets, high-definition (HD) maps, and platform APIs for ride-hailing and delivery. The HD maps provide detailed information about roads, which is essential for localization and driving decisions.

The company stated that this asset sharing would allow other industry players to develop autonomous driving technologies without having to independently build all the underlying infrastructure. Kakao Mobility also plans to share operational resources, including fleet management systems and on-site response capabilities, to support a nationally open autonomous driving ecosystem.

Concrete Example: The Autonomous Taxi Service in Gangnam

A concrete example of Kakao Mobility's commitment to autonomous driving is its nighttime autonomous vehicle service in the Gangnam district of Seoul. This service, available through the Kakao T platform, allows users to access autonomous driving services in addition to existing mobility options. Since its launch on September 26, 2024, until February 28, 2026, the nighttime autonomous taxi service in Gangnam recorded 7,754 rides with no accidents attributed to the autonomous driving technology, according to the Seoul Metropolitan Government. On average, about 24 rides are completed each day of operation.

The service, which started as a free pilot, became a paid operation in April 2026. The fleet has expanded from three to seven vehicles, excluding two reserve vehicles. Users can call the service via Kakao T using either the Seoul autonomous vehicle icon or the classic taxi booking menu. Kakao T consolidates several mobility services into a single application, including taxi, navigation, and vehicle-related services.