AI Adoption in Manufacturing: Insights from 100 Companies
Our analysis of the top 100 manufacturing companies by revenue from the Forbes Global 2000, spanning automotive, industrial equipment, chemicals, consumer electronics, and more across 15 countries, reveals two clear patterns in how manufacturers approach artificial intelligence.
Our analysis examines two key indicators of AI maturity:
- strategic partnerships, including startup investments and joint ventures,
- open-source contributions to AI projects.
Methodology
1. Data Collection
Partnership data came from three sources: company press releases, SEC filings, and industry databases. We searched for announcements between 2020 and 2025 that mentioned joint ventures, technology agreements, or research collaborations involving AI.
For open-source contributions, we searched GitHub for official company accounts and verified repositories with AI-related keywords. We tracked repository activity and licensing, and ensured the corporate affiliation was legitimate.
Every piece of data was verified across multiple sources. We also used an “AI Term Found” filter to separate real AI projects from generic “digital transformation” announcements that barely mention AI.
All data was verified through multiple sources with an “AI Term Found” classification to distinguish genuine AI initiatives from generic technology announcements.
2. Verification and Filtering
Companies announce “digital transformation initiatives” constantly. Most don’t involve AI at all. We built a filter that searched page text for specific terms: AI, machine learning, deep learning, data science, etc.
- Status code filtering: Only live pages (200 or 301) were accepted. Redirects were manually reviewed to confirm relevance.
- Keyword validation: If AI-related terms appeared more than once in the page body, the entry was marked as relevant; single mentions or domain-level terms (e.g., “.ai” in URLs) were ignored.
- Manual inspection: For companies with unclear results, researchers reviewed the text content directly to confirm whether the findings represented actual AI adoption or unrelated references.
This step-by-step filtering helped remove hallucinated or generic results such as sustainability reports mistakenly categorized as AI-related.
3. Metrics Used
Following multiple test runs and revisions, five metrics were finalized:
We started with five metrics, but only three gave us usable data.
- Partnerships and open-source contributions had concrete numbers we could track. Each partnership had a date, companies involved, and documentation. GitHub repositories showed commit history and activity levels. These made it straightforward to build charts.
- Employee training and use cases didn’t work out. Companies mention “AI training programs” in annual reports without explaining what employees actually learned.
- AI Adoption Output turned out to duplicate partnership data. A company announcing an “AI initiative” was usually just describing a partnership we’d already counted.
1.AI Partnerships
Manufacturers increasingly partner with major tech companies for AI collaboration.
Partnership Types
- Technology agreements make up 68% of partnerships. A typical agreement gives the manufacturer access to cloud AI services or pre-built models. General Motors uses Microsoft Azure for vehicle data processing. Toyota partnered with NVIDIA for autonomous driving simulation.
- Joint ventures account for 18%. These involve shared investment and longer timelines. HD Hyundai and Palantir built a data analytics platform for shipbuilding. CNH Industrial created a joint venture with several AI startups focused on autonomous farming equipment. Continental partnered with Horizon Robotics to manufacture AI chips for vehicles.
- Research partnerships represent 14%. Siemens collaborates with technical universities on industrial AI. Hitachi funds research labs working on predictive maintenance algorithms. BASF partners with AI institutes studying chemical process optimization.
Industry Patterns
- Automotive companies formed the most partnerships 32% of the total. General Motors and Hyundai Motor each have 10 partnerships. Autonomous driving requires integration across sensors, computing, and decision-making systems, which pushes companies to work with multiple AI providers.
- Industrial equipment manufacturers account for 28% of partnerships Caterpillar partners with cloud providers for equipment monitoring. Komatsu works with NVIDIA on construction site automation.
- Electronics and component makers represent 15%. These companies often partner to integrate AI into their products rather than their manufacturing processes.
Geographic Distribution
- Chinese manufacturers partner most frequently. CRRC has 8 partnerships, Great Wall Motor has 7, and Midea Group has 6. This reflects national policies pushing industrial digitalization.
- European companies average 3-4 partnerships each. They spread partnerships across cloud providers, research institutions, and specialized AI companies.
- North American manufacturers focus partnerships on major technology companies. Ford works primarily with Google Cloud, while Lockheed Martin partners with AWS and Palantir.
2. AI Open-Source Contributions
Manufacturers participate far less in open-source AI than in partnerships. Contributions mainly focus on industry-specific applications rather than general AI frameworks. This low participation rate contrasts sharply with the technology sector, where open-source contribution is standard practice for AI development.
Figure 2. Companies with AI Open-Source Contributions (2020–2025)
Contribution Patterns and Focus Areas
Industrial Applications Dominate Contributions
Companies contributing to open-source AI projects focus on industrial-specific applications rather than general AI frameworks. Siemens released open-source AI models for industrial anomaly detection, while Contemporary Amperex Technology published battery simulation toolkits with AI components. These contributions address manufacturing-specific problems not covered by general-purpose AI libraries.
Infrastructure and Data Processing Tools
Several manufacturers contribute to data processing infrastructure that supports AI applications. Caterpillar contributes to Apache Spark for industrial IoT data processing, while GE Vernova contributed machine learning models for energy forecasting to Apache PredictionIO. These contributions focus on the data pipeline components that enable AI rather than core AI algorithms.
Computer Vision Applications
Automotive manufacturers show the highest open-source activity in computer vision projects. BMW Group released open-source AI tools for autonomous driving simulation, while Bridgestone contributed tire analysis algorithms to the OpenCV project. This concentration reflects the sector’s focus on visual perception systems for autonomous vehicles.
Highlights
- Siemens released open-source models for detecting anomalies in industrial equipment. The models identify unusual patterns in sensor data from manufacturing lines.
- BMW published simulation tools for testing autonomous driving algorithms. The toolkit lets researchers test perception systems without physical vehicles.
- Caterpillar contributes to Apache Spark, specifically the modules that process IoT sensor data from heavy equipment.
- Contemporary Amperex Technology (CATL) published battery simulation software with AI components. The models predict battery degradation under different usage patterns.
- GE Vernova contributed energy forecasting models to Apache PredictionIO. The models predict electricity demand using weather data and historical patterns.
3. AI Adoption Output
This metric captured the number of identifiable AI initiatives per company.
The findings showed a concentration among a few leaders: Siemens (9 initiatives), GE Vernova (4), and several others with one or two projects. These initiatives include AI-enabled products, predictive maintenance systems, and automation tools.
Figure 3. Number of AI Initiatives by Company (Top 10)
Key Findings
Partnerships Are the Default Strategy
- 89 companies formed at least one AI partnership since 2020. Microsoft, NVIDIA, and Huawei appear most frequently—together accounting for 30% of all partnerships.
- Most manufacturers skip internal AI development and go straight to established providers. The logic is straightforward: cloud platforms deploy faster than building in-house capabilities.
Open-Source Barely Registers
- Only 7 companies contribute to public AI projects. The gap is stark 12 times more manufacturers partner than share code.
- When they do contribute, it’s industry-specific tools: battery simulation, tire analysis, equipment monitoring. Nobody’s publishing general-purpose AI models or contributing to frameworks like TensorFlow.
Investment Levels Vary Wildly
- Siemens leads with 9 AI initiatives. GE Vernova and KIA follow with fewer but consistent projects. Most other companies have 1-2 partnerships and nothing else.
- This suggests manufacturers are testing AI rather than committing to it. A couple of pilot projects doesn’t indicate deep investment.
Automotive and Industrial Equipment Pull Ahead
- Automotive represents 32% of partnerships. General Motors and Hyundai Motor both have 10 partnerships each—the highest counts.
- Autonomous driving creates pressure to integrate multiple AI systems: computer vision, sensor processing, decision algorithms, simulation. Companies can’t build all of this alone.
- Industrial equipment manufacturers follow similar patterns. Caterpillar and Komatsu need AI for equipment monitoring and automation.
- Chemical and materials producers move slower. BASF has research partnerships, but most chemical manufacturers show minimal AI activity.
Closed Innovation Wins
Manufacturers buy AI through partnerships but keep their implementations private. They’ll license Microsoft Azure but won’t publish their predictive maintenance algorithms.
Furher Reading
- Generative AI in Manufacturing: Use Cases & Benefits
- Manufacturing AI: Top 15 tools & 13 real life use cases
Cem's work has been cited by leading global publications including Business Insider, Forbes, Washington Post, global firms like Deloitte, HPE and NGOs like World Economic Forum and supranational organizations like European Commission. You can see more reputable companies and resources that referenced AIMultiple.
Throughout his career, Cem served as a tech consultant, tech buyer and tech entrepreneur. He advised enterprises on their technology decisions at McKinsey & Company and Altman Solon for more than a decade. He also published a McKinsey report on digitalization.
He led technology strategy and procurement of a telco while reporting to the CEO. He has also led commercial growth of deep tech company Hypatos that reached a 7 digit annual recurring revenue and a 9 digit valuation from 0 within 2 years. Cem's work in Hypatos was covered by leading technology publications like TechCrunch and Business Insider.
Cem regularly speaks at international technology conferences. He graduated from Bogazici University as a computer engineer and holds an MBA from Columbia Business School.
Be the first to comment
Your email address will not be published. All fields are required.