During the 2026 sessions of the National People's Congress, a proposal by Lei Jun—founder of Xiaomi Group and a deputy—has once again turned the auto industry's focus back to "people."
In a proposal titled "Accelerating the Cultivation of Composite Talent for Intelligent Connected New Energy Vehicles," Lei Jun points out that China's gap in such talent has hit 1 million. Multiple sources suggest a fierce "war for talent" is unfolding in the intelligent connected vehicle sector.
Yet, there is another side to the coin: the domestic market is gripped by "hyper-competition" and reshaping. The entire industry faces the dilemma of rising revenue without rising profit, with news of plant closures, layoffs, or cuts to non-core businesses surfacing regularly.
This structural contradiction—slashing traditional redundant staff while struggling to find talent even at high pay—reveals a stark reality: China's auto industry has entered the "deep water" zone driven by artificial intelligence, but the "talent foundation" needed to sustain this transformation remains unfinished.
The Shortage is High-End Talent
Does the auto industry really lack people? If looking at total labor, the answer is no. But for the "intelligent connected new energy" market Lei Jun emphasizes, the answer is a resounding yes—and the situation is severe.
Traditional auto manufacturing focused on hardware, with core competitiveness rooted in mechanical stability and performance. But under the "AI+" wave, the car's essence is shifting from a transport tool to a mobile intelligent terminal and a node in digital space.
This shift has directly caused a fracture in talent demand. Public data shows that in the first three quarters of 2025, the penetration rate of L2 driver assistance in domestic passenger vehicles exceeded 50%, while L3 and L4 autonomous driving have entered large-scale pilot phases.
Image source: Chery Holdings Group
As software's share of value in cars climbs, automakers no longer need single-discipline experts obsessed with internal combustion or chassis tuning. They need cross-boundary players who understand the logic of "software-defined vehicles."
The 1 million-person gap Lei Jun mentions refers to talent with "composite, cross-disciplinary" capabilities. These individuals are no longer confined to a single subject but possess a multi-dimensional competitive model:
They must grasp the engineering value of AI algorithms while adopting a product manager's perspective to translate complex technology into actionable smart cockpit solutions.
And they need cross-domain collaboration skills. Intelligent connected vehicles involve computer science, AI, big data, and communications. Talent must break down "technical silos" to collaborate deeply across perception, decision-making, and control teams. They also need systems thinking—viewing the vehicle's intelligent architecture holistically rather than just optimizing a single module.
Currently, this "AI-savvy, auto-savvy, scenario-savvy" composite talent is extremely scarce. Supply and demand are imbalanced, leaving many companies in the awkward position of offering high salaries yet struggling to find the right leaders.
Partnering with Universities
Addressing this disconnect, Lei Jun's proposal offers a systematic solution ranging from discipline construction to industry-education integration. This echoes the explorations already underway between leading automakers and universities.
First, reconstructing the knowledge system by establishing a first-level discipline. Lei Jun's primary suggestion is a "top-level design" change. He proposes creating "Intelligent Electric Vehicles" as a first-level interdisciplinary discipline and including it in the national catalog. This would grant the field independent academic status and resource allocation, fundamentally changing the current reality where talent training is scattered across mechanical, electronic, and computer science departments.
Image source: Geely Holding Group
Then, shifting from "professional education" to "cross-boundary cultivation." Lei Jun advocates for a "dual mentor" and "combat-oriented" training model, guided by actual industry needs.
Universities must drive cross-department teaching. For instance, automotive engineering should partner with computer science, allowing students to master core electronic system technologies while studying semiconductors, and dive straight into autonomous driving scenarios when learning large AI models.
In disciplines like autonomous driving, students should encounter the integrated application of computer vision, machine learning, and control theory during their studies, solving real perception and decision-making problems through project-based learning.
Practically, Lei Jun suggests building a new mechanism for collaborative industry education, encouraging companies to invest in talent development. In fact, pioneers already exist. Automakers like Geely and Volkswagen have long tried to shorten the conversion cycle for graduates from "knowledge reserve" to "composite application" by opening specialized colleges and co-building industry-education bases with universities.
In this process, companies can use a "mentor system + project combat + frontier training" approach, where senior engineers guide students through real testing, R&D, and marketing. For example, during their junior and senior years, students could enter off-campus corporate bases to perform hands-on work on specific smart control or algorithm optimization tasks, truly combining "classroom theory with actual operation."
To spur corporate enthusiasm, Lei Jun also proposes including participation in discipline construction as a criterion for policy support and excellence awards. Meanwhile, experts suggest promoting certification for "industry-education integrated enterprises" and perfecting the evaluation system for skilled talent, giving those who "understand technology, can manage, and can execute" broader career advancement channels.
The landing of "AI+" in the auto industry is, in essence, a relay race centered on "people."
As industry insiders note, composite talent is the key connecting technology with business, and innovation with implementation. The current gap of 1 million is both a pain point of industrial transformation and a reservoir of future potential. By establishing cross-disciplinary systems, deepening school-enterprise cooperation, and increasing tech innovation training, China is attempting to build a more flexible and efficient mechanism for talent supply.
