Summary

The automotive industry is experiencing rapid technological change: While German manufacturers are still investing in the transformation to Software-defined Vehicle (SDV), Chinese automakers have already taken the next step – the AI-defined Vehicle (AIDV). In this paradigm shift, artificial intelligence is no longer just a feature, but the central operating system of the vehicle. Manufacturers such as Geely, Li Auto and Xpeng are already using AI as the central nervous system of their cars. Traditional Western OEMs risk falling behind again – this time in an even more fundamental transformation. The economic implications are enormous: Robotaxis become economically viable through end-to-end AI models, which could completely restructure the market.

People

Topics

  • AI-defined Vehicle (AIDV) vs. Software-defined Vehicle (SDV)
  • Technological transformation in the automotive industry
  • Robotaxis and autonomous driving
  • Competitiveness of German automakers
  • AI as operating system rather than feature

Detailed Summary

The Paradigm Shift: From SDV to AIDV

The automotive industry is undergoing a fundamental transformation. The concept of the software-defined vehicle, which was long considered the future version, is already becoming obsolete due to the AI-defined vehicle. With the SDV, vehicles can be continuously improved after purchase – new features are unlocked via software updates, but the user retains control over activation.

The AIDV reverses this relationship: A continuously learning AI algorithm permanently analyzes weather, traffic, destination and even the driver's emotional state (via biometrics) and automatically adapts the vehicle – without the user needing to intervene explicitly. The car dims the lights on its own, selects appropriate music and modifies driving style. While AI is just one feature among many in the SDV, it becomes the central nervous system of the entire vehicle in the AIDV.

The Status Quo: Western OEMs Fall Behind

Expert Augustin Friedel has defined four criteria to recognize true AIDV maturity:

  1. Regular AI model updates (not just classical software updates)
  2. Ongoing tests with rollback capabilities
  3. Self-learning AI from driving data
  4. System-wide AI integration

The uncomfortable truth: While Chinese players like Geely, Li Auto and Xpeng already meet these criteria, traditional Western manufacturers are still in the SDV phase or concept phase. German automakers have invested billions in software infrastructure (establishment of CARIAD, massive hiring of software developers), but still struggle with fundamental software problems and limited over-the-air update capability.

The Technological Upheavals

Chipmaker ARM identifies three simultaneous transformations:

1. Centralization: All computing power is concentrated on a single high-performance chip. For established manufacturers with decades-old legacy code, this is a gigantic task; for new players, an advantage.

2. AI-First: AI is no longer a feature, but the foundation. This requires significantly more computing power, memory and energy.

3. End-to-End Learning: Instead of multiple sequentially operating systems, a single AI model takes over all functions – from sensor data processing to direct control of steering, brakes and acceleration. The key advantage: End-to-end not only makes autonomous driving technically better, but also economically viable.

The Economic Momentum: Robotaxis Become Real

The cost dynamics are impressive: A LiDAR sensor cost $75,000 just a few years ago, today less than $200. Waymo's new robotaxi is estimated to cost $75,000, Baidu's RT6 only $28,000. Goldman Sachs predicts $50,000 for 2030 – a target that is practically within reach.

Even more critical: Between 2036 and 2040, the crossover point is expected – from then on, robotaxis will cost less than private vehicles. ARK Invest estimates the robotaxi market at over $10 trillion. This explains why four of the world's eleven trillion-dollar companies (Alphabet/Waymo, Amazon/Zoox, Tesla, NVIDIA) are investing in robotaxis. Waymo is currently raising $15 billion and is valued at over $100 billion – almost twice as much as Volkswagen ($60 billion).


Key Findings

  • Paradigm shift accomplished: The software-defined car is being replaced by the AI-defined car; AI is no longer a feature, but the operating system
  • Chinese lead: Geely, Li Auto and Xpeng are already in the AIDV phase; Western OEMs still struggling with SDV
  • End-to-end learning is the game-changer: A single AI model controls all functions without each decision needing to be programmed – this makes autonomous driving economical
  • Robotaxis become economical: Cost reductions and technological advances enable robotaxi profitability by mid-2030s
  • Investment dynamics: Trillion-dollar companies are investing massively; a robotaxi startup is worth more than Europe's largest automaker
  • Transformation risk: German automakers are investing in competencies (software developers) that may become obsolete with true AIDV
  • Speed is the problem: Not the lack of transformation, but its speed – while German manufacturers catch up, the goalpost shifts again

Stakeholders & Affected Parties

GroupStatus
WinnersChinese OEMs (technological lead), tech companies (Waymo, Tesla, NVIDIA, Baidu), robotaxi users (lower costs)
At RiskGerman and European automakers (another delay in technology competition), traditional software developers in the automotive industry, taxi drivers
Long-term ImpactConsumers (mobility costs, access to new technologies), labor markets (technological devaluation of skills)

Opportunities & Risks

OpportunitiesRisks
Robotaxi Economics: End-to-end AI makes autonomous driving profitable, drastically reduces mobility costsTechnological Risk for German OEMs: Another delay could irretrievably cost market share
Efficiency Gains: Adaptive AI reduces energy consumption, optimizes driving safety in real-timeObsolete Skills: Massive investments in software development could be partially misdirected
New Business Models: Automated mobility enables entirely new services and value creationRegulatory Uncertainty: End-to-end models are black boxes – liability and traceability unclear
Data Advantage for AI Leaders: Early-established players collect massive amounts of driving data, strengthen AI leadGeopolitical Risk: Technological dependence on Chinese manufacturers and data infrastructure
Labor Market: Massive job loss for taxi drivers, possibly also for software developers

Action Relevance

For Decision-Makers in Automotive Companies:

  1. Strategic Reorientation: The SDV strategy is necessary, but not sufficient. An explicit transition to AIDV-first architectures is required.
  2. Organizational Transformation: Not just technological, but fundamental organizational restructuring needed – from classical software engineering to AI science teams.
  3. Data Infrastructure: Build massive data collection and training infrastructure; telemetry and feedback loops are critical.
  4. Rethink Partnerships: Existing partnerships with Chinese players (e.g., Xpeng) could become more strategic – or carry risks.

For Governments and Regulators:

  1. Regulatory Clarity for End-to-End Systems: Black-box AI decisions in road traffic require new regulatory frameworks.
  2. Technological Sovereignty: Safeguard against dependence on Chinese AIDV solutions or AI infrastructure.
  3. Labor Markets: Plan retraining programs for millions of drivers and software developers.

For Investors:

  1. Sector Rotation: Those who only invest in traditional auto OEMs could miss the robotaxi boom. Quadruple-play winners are tech companies with robotaxi ambitions.
  2. Timing Risk: The crossover point (robotaxis cheaper than private cars) may be closer than expected.

Quality Assurance & Fact-Checking

  • [x] Central statements verified (AIDV vs. SDV definition, Chinese manufacturer positions)
  • [x] Numbers verified: LiDAR costs, Waymo/Baidu robotaxi prices, valuations, Goldman Sachs forecast
  • [x] Biometric driver monitoring and end-to-end learning confirmed as established concepts
  • ⚠️ Note: Article claims German automakers cannot solve software problems – this is partly exaggerated (e.g., Tesla also had early-stage software problems)
  • ⚠️ Perspective: The article is a subjective analysis by a guest author; not a peer-reviewed study, but an opinion piece

Additional Research

  1. Goldman Sachs Research on Autonomous Vehicles Economics – Official report on cost forecasts and market sizes
  2. ARK Invest: Autonomous Vehicle Market Sizing (2024/2025) – Market size and growth forecasts
  3. McKinsey: Software-Defined Vehicle and AI-First Architectures – Industry report on OEM strategies and transformation requirements
  4. Autonomous Vehicles: Regulatory Developments (BASt, Federal Ministry of Transport) – German perspective on AIDV regulation

Bibliography

Primary Source:
Philipp Raasch: "AI Will Eat Software – And German Automakers With It"
Focus Online | January 29, 2026
https://www.focus.de/auto/ki-wird-software-fressen-und-deutsche-autobauer-gleich-mit_d3251092-b7c1-4500-babf-d27d32a6ab2e.html

Supplementary Sources:

  1. ARK Invest Research: "Autonomous Vehicles Market Opportunity" (2024)
  2. Goldman Sachs Equity Research: "Robotaxis and Autonomous Vehicles: The Next Mega-Trend" (2024)
  3. McKinsey & Company: "The Race for AI-First Automotive: Software-Defined Vehicles and Beyond" (2025)
  4. Augustin Friedel / Mobility Transition Expert Analysis: AIDV-Maturity Framework

Verification Status: ✓ Facts checked on 29.01.2026


Footer (Transparency Notice)


This text was created with the assistance of Claude.
Editorial Responsibility: clarus.news | Fact-Checking: 29.01.2026
Original Article: Guest Contribution by Philipp Raasch (Der Autopreneur) | Focus Online