China L3 Autonomous Driving

The 5 Numbers MIIT Did Not Disclose About Its First L3 Approvals

Apr 25, 2026 EngiVolt Pro

// DATA GAP AUDIT — China L3 Approvals

On December 15, 2025, China’s Ministry of Industry and Information Technology granted the country’s first L3 production approvals. The approval extends to two cars in two cities. Industry coverage compressed the news into “L3 has landed in China” — a claim with five missing numbers under it. Each missing number is a question any audit must ask before evaluating the regulatory framework. As of April 25, 2026, none of the five has been disclosed by MIIT, the Beijing Municipal Government, the Chongqing Municipal Government, or the manufacturer itself.

The Five Missing Numbers

The MIIT approval and accompanying municipal pilot frameworks specify that L3 mode operates within a defined operational design domain. The ODD is bounded by speed (50 to 80 km/h depending on configuration), lane configuration (single lane), and geographic scope. The geographic scope is where the disclosure stops. The total approved road kilometers, the specific roads, the identity of approved operating entities, the public retail availability timeline, and the retail trim pricing — five separate variables — are absent from every public MIIT, municipal, or manufacturer document we reviewed for this audit.

MIIT AND MUNICIPAL AUTHORITIES — DATA GAPS: WITHHELD BY OEM
[?]GAP-W17-01
Total approved L3 road length -- Beijing and Chongqing
IMPACT: Without total approved kilometers, the marketing claim L3 has landed cannot be quantitatively verified against the 5,490,400 km national road network. Conservative upper bound (100 km combined) puts the approved footprint at under 0.002 percent of national roads.
BEIJING AND CHONGQING TRANSPORT AUTHORITIES — DATA GAPS: WITHHELD BY OEM
[?]GAP-W17-02
Specific approved L3 roads or expressway segments
IMPACT: A buyer cannot determine whether L3 mode applies to their commute without the specific road list. The information exists internally for the designated operating entities but is not in any public document.
MIIT AND MUNICIPAL AUTHORITIES — DATA GAPS: WITHHELD BY OEM
[?]GAP-W17-03
Identity of approved operating entities
IMPACT: The approval extends to designated operating entities, not retail consumers. Which fleet operators or commercial mobility services are the designated entities is not disclosed. Fleet buyers seeking partnership cannot identify counter-parties.
CHANGAN, BAIC, MIIT — DATA GAPS: WITHHELD BY OEM
[?]GAP-W17-04
Public retail availability timeline for the L3 trim
IMPACT: Retail consumers cannot activate L3 mode through private ownership under the December 2025 approval. When that restriction lifts -- if it does -- is not in any public timeline.
CHANGAN AND BAIC — DATA GAPS: WITHHELD BY OEM
[?]GAP-W17-05
Retail trim pricing for L3-approved configurations
IMPACT: The premium for L3-capable trims relative to baseline configurations is not disclosed for the two MIIT-approved configurations. Buyers cannot compare value across competing trims without published pricing.

Why Three of Five Should Be Public Already

The total approved road length, the specific approved roads, and the identity of approved operating entities are not subject to the same competitive-disclosure constraints as retail pricing or product timelines. They are regulatory facts about a public policy decision. In comparable jurisdictions — the EU’s UNECE R157 framework, California’s autonomous vehicle deployment program, Japan’s Road Traffic Act amendments — equivalent ODD definitions and operating entity rosters are publicly disclosed at the regulator level. The Chinese non-disclosure is itself a regulatory choice. The choice is consistent with the framework’s “controlled commercialization pathway” language, but it is also the audit gap that prevents independent verification.

Until the five numbers are published, “L3 has landed” remains a marketing statement about a regulatory event whose substantive scope cannot be quantitatively measured. That distinction matters for any buyer, fleet operator, or analyst pricing 2027 commercialization timelines.

Tracking

Monitoring Trigger W17-T3 (Beijing or Chongqing publishes ODD road length) tracks GAP-W17-01 and GAP-W17-02. Monitoring Trigger W17-T1 (second batch of MIIT L3 production approvals, expected late 2026) may close GAP-W17-04 if the second-batch announcement specifies retail timeline. EngiVolt Pro will publish a Post-Audit Note when any of the five numbers is publicly disclosed.

// Get the Report on Patreon

Full Engineering Audit Report

37-page report (English plus Chinese), 5 Challenge Points, 44 sourced claims, complete monitoring triggers schedule through Q3 2027.

49 USD — Get the Report →