仅20%功能实现盈利！车载AI看似遍地黄金 车企为何越投入越亏钱？

This article highlights the current paradox of the in-vehicle AI industry: it is a booming sector mired in widespread unprofitability. Key takeaways are as follows:

1. Core industry overview: Intelligence has become the core competitive focus of the auto industry’s second half, and is widely seen as the key growth driver for the sector over the next decade. The global in-vehicle AI market is projected to exceed $50 billion by 2034, representing a nearly fourfold increase over 10 years. However, currently only 20% of in-vehicle AI functions achieve positive profitability, with 80% either breaking even or losing money. Most automakers are pouring massive resources into R&D but struggle to turn a profit.

2. Root causes of unprofitability: The traditional one-time capital investment model of legacy automakers is completely mismatched with the ongoing cloud computing and operation and maintenance costs that in-vehicle AI requires. In addition, most automakers blindly measure智能化 level by the number of AI functions offered, without conducting granular cost-benefit analysis. This has led to a large number of "zombie functions" with less than 5% daily active usage and underperforming low-value features, which continuously drag down profitability.

3. Practical paths to profitability: Automakers should cut underperforming loss-making functions promptly, adopt a hybrid local computing-cloud AI architecture to cut costs while improving user experience, overhaul monetization models, and shift to a value-based on-demand payment model.

This article delivers actionable insights for automotive brands on the current state of the in-vehicle AI industry and its profitability challenges. Key takeaways are as follows:

1. Industry and consumer trends: Intelligence is now the core competitive advantage for automotive brands, and the in-vehicle AI market offers enormous growth potential, with a projected nearly fourfold expansion over 10 years, making it a widely recognized high-growth opportunity. However, the entire industry currently struggles with high input costs and low returns, with only 20% of AI functions turning a profit.

2. Product R&D insights: Do not blindly stack AI functions to compete, and do not use function count as a measure of intelligence. Instead, build a granular per-function cost-benefit accounting system, promptly eliminate unprofitable zombie and low-value functions, and focus on refining core high-frequency features that meet real user needs.

3. Brand marketing and commercialization insights: Do not treat in-vehicle AI as just a marketing gimmick to dress up products; it should be built as a long-term, finely operated value-added business. The traditional flat subscription model has a low ceiling, so brands should transition to building active transactional intelligent agents, expand value-added service scenarios, and adopt the new model of value-based on-demand payment. At the same time, the hybrid local-cloud architecture can be adopted to reduce costs while improving user experience.

The in-vehicle AI track offers broad growth opportunities but also carries clear profitability risks. Key takeaways are as follows:

1. Market growth and opportunities: Intelligence is the core competitive direction of the auto industry’s second half, and in-vehicle AI is the key growth driver for the next decade. The global market will exceed $50 billion in 10 years, and the capital market remains bullish on the sector. Vehicles with AI upgrades have greater room for market growth.

2. Risk warnings: The industry is currently riddled with widespread profitability traps: 80% of in-vehicle AI functions are either unprofitable or just break even. Blindly promoting vehicles with a surplus of AI functions will raise costs and compress profit margins, and poor user experience from redundant features can also trigger negative user reviews, hurting sales and brand reputation.

3. Opportunites and best practices: During product selection and promotion, prioritize core AI functions that are high-frequency, practical, and deliver a great user experience. Focus marketing messaging on practicality rather than function count. You can also partner with new business models built around active transactional intelligent agents to unlock new revenue from value-added travel services, and avoid the trap of function-based competition.

The profitability crisis and ongoing transformation of the in-vehicle AI industry bring multiple insights for automotive manufacturing factories. Key takeaways are as follows:

1. Changes to production and design requirements: Users’ core demand for in-vehicle AI is practicality and good experience, not a large quantity of features. When supporting the production of in-vehicle AI related hardware, factories do not need to allocate redundant hardware for unnecessary low-value features. Focusing production support on core high-frequency functions can effectively reduce production costs and better match market demand.

2. New business opportunities: The most widely industry-recognized technical solution today is the hybrid local computing-cloud intelligence architecture. This solution significantly increases demand for high-performance automotive-grade chips, bringing new incremental order opportunities for factories that produce automotive chips and supporting hardware.

3. Insights for digital transformation: In-vehicle AI is an ongoing operation business, which is completely different from traditional one-time delivered hardware. Factories need to adapt to industry changes, move past the traditional heavy-asset one-time production mindset, and build a cost management system adapted for AI operations to better serve automaker clients and improve their own market competitiveness.

The current state of the in-vehicle AI industry and its ongoing transformation toward profitability point to new business directions for in-vehicle related service providers. Key takeaways are as follows:

1. Industry development trends: The old growth model of blindly stacking AI functions has reached its end. The industry’s core future focus will shift to fine-grained operation and commercial monetization. Service providers should no longer position function count as their core selling point, and instead shift their focus to improving user experience and helping clients achieve profitability.

2. Core client pain points: The top pain points for automaker clients today are the high ongoing operating costs of in-vehicle AI and difficulties with commercial monetization: a large share of functions fail to turn a profit after investment, while the industry as a whole struggles with poor user experience and low function utilization. All of these are areas service providers can enter.

3. Solution directions: Service providers can prioritize developing and promoting the hybrid local-cloud AI architecture to help automakers cut operating costs and improve user experience. In addition, they can develop related technology and operation solutions centered on active transactional intelligent agents, helping automakers connect to full-scenario value-added travel services, open up new revenue streams, and adapt to the new on-demand value-based payment business model.

The shift toward profitability in the in-vehicle AI industry brings new development directions for in-vehicle related platform operators. Key takeaways are as follows:

1. New industry demands for platforms: The in-vehicle AI industry has shifted from function-stacking to exploring commercial monetization. Platforms need to integrate full-category value-added travel service resources to support the new scenario of active transactional intelligent agents, and meet automakers’ demand to connect to services such as in-car ordering, charging station reservation, car washing and parking.

2. Platform merchant recruitment and operation directions: During recruitment, platforms should prioritize onboarding AI technology service providers and value-added travel service suppliers that can improve AI experience and help automakers achieve monetization, and phase out providers that only deliver low-quality redundant functions to optimize the platform’s service structure.

3. Risk mitigation points: Platforms should avoid the boom-time trap of blindly expanding function count, and should not use function count as a core selling point. Instead, they should focus on helping automakers control operating costs and improve monetization efficiency, build a finely tuned service operation system, avoid the widespread industry profitability risk, and capture new opportunities from industry transformation.

The in-vehicle AI industry is facing new developments and prominent contradictions, which open up new directions for research. Key insights are as follows:

1. New industry problems: The industry currently faces a clear development paradox: it is widely favored by both capital and industry insiders, with a projected nearly fourfold market expansion over 10 years and enormous growth room, but in actual operation only 20% of AI functions achieve profitability. The entire industry is trapped in a high-investment, low-return dilemma. The core contradiction lies in the fundamental mismatch between AI’s continuous operation requirement and the legacy automaker’s one-time investment business model, combined with the lack of a proper cost accounting system and blind function stacking, which ultimately leads to the profitability crisis.

2. New industry trends: The competitive logic of in-vehicle AI has already been rewritten. The粗放烧钱扩张 model is gradually coming to an end, and new business models and technical roadmaps are emerging. The technical focus is shifting to hybrid local-cloud architecture, the business model is shifting from fixed subscription to value-based on-demand payment, and function positioning is shifting from passive assistance to active transactional intelligent agent.

3. Future research can focus on the implementation of new business models, the construction of a cost accounting system adapted to in-vehicle AI, and the balance path between technology and business, to explore feasible solutions for the industry to break out of its current predicament.