Introduction:
On April 16, 2025, a low-profile announcement from the First Division of Equipment Industry at China’s Ministry of Industry and Information Technology (MIIT) quietly signaled a turning point for the nation’s intelligent connected vehicle industry. The convening of this meeting—centered on product admission and software update management—marked the industry’s transition from “tech frenzy” to a new phase of “safety-driven navigation.” As penetration of L2+ advanced driver-assistance systems (ADAS) is expected to surpass 65%, the question of how to strike a balance between innovation and safety has become one that every link in the automotive chain must now answer.
Policy Direction: Intelligent Vehicles Have New “Traffic Rules”
The MIIT’s latest initiative introduces systematic regulations throughout the full product life cycle:
- Admission test upgrades: 39 new scenario-based assessment criteria have been added, including recognition in low-light nighttime conditions and dynamic obstacle avoidance in construction zones. The threshold for simulation testing mileage has been raised from 10,000 km to 50,000 km.
- Closed-loop data regulation: Automakers are now required to build cloud-based data management platforms, capable of real-time uploads of 12 categories of data, such as vehicle operating status and system decision logs. Data must be retained for no less than 10 years.
- OTA transparency: Before any over-the-air software update, automakers must disclose a full list of functional changes. Any updates involving driving strategy adjustments must be accompanied by a third-party validation report.
- Responsibility system reconstruction: Car manufacturers are clearly defined as the primary technical entity responsible during periods when autonomous driving functions are activated. Insurance providers may dynamically adjust premium coefficients based on automakers’ safety ratings.
Of particular note is the policy’s emphasis on “gradual implementation.” Currently sold models are given a 6-month buffer period for rectification, while newly declared models must immediately comply with the new standards. This “old cars, old rules; new cars, new rules” transitional approach provides room for adjustment while ensuring the safety floor of tech iteration is upheld.
Technological Evolution: From “Feature Stacking” to “Reliable Redundancy”
Driven by policy, the industry’s technological roadmap is showing three notable shifts:
- Multimodal perception systems: Installation rates for LiDAR have surged from 18% in 2024 to 47%, with millimeter-wave radar and vision fusion solutions becoming the mainstream configuration.
- Conservative decision logic: One automaker has raised the safety buffer for lane-change decisions from 0.8 seconds to 1.5 seconds—preferring to sacrifice efficiency to lower risk.
- Standardized validation systems: Leading companies have jointly established China’s first intelligent driving scenario database, covering 23,000 locally distinctive conditions including rural narrow roads and multi-level urban traffic.
According to third-party inspection agencies, newly declared models have improved their pass rates by an average of 21% on key metrics such as pedestrian crossing and foreign object recognition. Some companies have begun introducing aviation-grade “fail-operational” design principles, ensuring vehicles can maintain baseline safety performance even when a single system fails.
Consumer Guide: Understanding the Safety Logic Behind “Smart” Cars
For consumers, choosing an intelligent vehicle now involves paying attention to three new dimensions:
- Check certifications: Look for the “Intelligent Connected Vehicle Admission Mark” sticker on the vehicle body. The digital code can be used to check that model’s test scores on the MIIT’s official website.
- Read the data: Via automakers’ apps, consumers can view metrics such as cumulative intelligent driving mileage and system intervention frequency. For mature systems, the monthly manual takeover rate should be below 0.5%.
- Examine the architecture: Give priority to models with domain-centralized electronic and electrical architecture, which offers 3x greater stability in software updates and fault isolation compared to distributed architecture.
Industry experts recommend referring to the “Intelligent Driving System Maturity Index” published by China Automotive Engineering Research Institute (China Automotive Technology & Research Center), which aggregates road tests, simulation data, and user feedback across 12 dimensions and updates quarterly rankings by model.
Future Vision: Building an Integrated “Vehicle-Road-Cloud” Ecosystem
The deeper signal emerging from the policy meeting is that China’s intelligent connected vehicles are evolving from “standalone intelligence” to an integrated system ecosystem:
- Smart road transformation: Five pilot provinces have launched “smart highway” construction, where roadside infrastructure can push real-time risk alerts to vehicles.
- Cloud-based monitoring platform: The National Intelligent Vehicle Big Data Center now connects to over 4 million vehicles, enabling cloud-based safety risk prediction.
- Insurance innovation pilots: Shenzhen and Shanghai have launched user-based insurance (UBI) trials driven by vehicle data. Safe driving behavior could reduce premiums by up to 60%.
As one expert participant noted: “When every vehicle is both a producer of data and a beneficiary of the ecosystem, only then can a true safety net for intelligent mobility be woven.” What began as a change in product admission rules is now reshaping the foundational logic of the entire mobility ecosystem.
The day after the MIIT’s announcement, a certain automaker quietly replaced its showroom slogan—from “Free Your Hands” to “Protect Your Journey.” This subtle detail may best capture the industry’s transformation: When technological innovation returns to the essence of safety, only then can China’s intelligent vehicles truly set course for the stars and the sea.
[Disclaimer]: The above content reflects analysis of publicly available information, expert insights, and BCC research. It does not constitute investment advice. BCC is not responsible for any losses resulting from reliance on the views expressed herein. Investors should exercise caution.