Socar与Apex Mobility拟将车队实路与事故数据商品化供应自动驾驶AI训练市场

Socar (403550.KS) and Apex Mobility are making a full push into the self-driving data business. The plan is to process real-world driving and accident data — accumulated over 15 years of running a car-sharing service — into material for training artificial intelligence, and to supply it to the global autonomous-driving market.
Socar and Apex Mobility said on the 24th that they will take part in the 2026 Autonomous Mobility Industry Expo (AME 2026), a three-day event running from the 25th to the 27th at COEX in Seoul. Apex Mobility is a self-driving joint venture that Socar established in May with 150 billion won ($108 million).
At the expo, the two companies will put forward the concept of Korea's largest self-driving data platform. As autonomous-driving technology shifts rapidly from a rule-based approach built on high-definition maps to an end-to-end (E2E) method that processes multi-sensor data through a single AI model, their strategy is to secure an early lead in the supply of high-quality driving data, which has emerged as a decisive battleground in the market.
Socar collects more than 100 types of vehicle data in real time — including front and rear dashcam footage, GPS and steering angle — through some 25,000 connected cars operating nationwide. Its daily driving distance alone reaches about 1.1 million kilometers, 10 times the length of the country's road network, covering public-road conditions across all four seasons.
Its particular strength lies in its competitiveness in "edge case" data — unexpected situations — regarded as central to advancing self-driving AI. Socar collects more than 40,000 accident videos a year, along with ego-motion data (estimates of a vehicle's three-dimensional movement) at the moment of an accident, and has amassed more than 220,000 pieces of such specialized data to date. This data is refined into training data that meets international standards through Apex Mobility's de-identification, time-sync and vision-language-model (VLM)-based automatic labeling technologies.
The expo will also mark the first public unveiling of an Ioniq 5 prototype for self-driving data collection, built in-house by Apex Mobility. Equipped with a full sensor kit including a roof-mounted LiDAR and seven multi-view cameras, the vehicle will be used to supply custom-built datasets tailored to the regions, vehicle types and scenarios that clients request.
Park Jae-uk, CEO of Socar and Apex Mobility, said, "In the era of E2E autonomous driving, where AI directly handles perception and judgment, competitiveness depends on the quantity and quality of data secured on real roads." He added, "Based on our unrivaled real-time data pipeline, we will move ahead in earnest with supply partnerships with domestic and overseas automakers and self-driving companies."