Every time a BMW A driver remotely locks their vehicle, checks the battery level, receives a navigation update, or obtains a software fix over the air, which is considered a request.
With 24.5 million connected vehicles, these small interactions are adding up quickly. BMW’s fleet now produces more than 16.6 billion requests per day, processes 184 terabytes of data and 100 million API calls with sub-second latency, according to AWS I mentioned.
BMW now runs more than 600 AI use cases across the company He said. Engineers use artificial intelligence to perform crash simulations without building physical prototypes. Procurement teams use it to analyze supplier contracts and create tender documents. Factory systems use it to inspect welds in real time and report defects before the order moves down the assembly line.
It all runs on a common enterprise platform. The platform allows internal teams, including non-technical specialists such as battery engineers and logistics planners, to build and deploy their own AI tools without writing infrastructure code. More than 12,000 developers within the BMW Software Factory work on AWS, the company male. BMW is also using AI to perform automated root cause analysis of cloud outages, reducing incident diagnosis from hours to minutes, according to AWS. I mentioned. The system correctly identifies the root cause in 85% of cases.
The BMW factory floor detects defects and moves parts without human intervention
Before BMW Built Its Connected AI Platform on AWS The team behind its Intelligent Personal Assistant — the in-car system that learns driver preferences and suggests relevant features on the road — had to wait overnight for the model to complete its training.
Now, the platform runs on Amazon Elastic Kubernetes Service and distributes compute work across multiple GPUs simultaneously instead of processing sequentially on a single machine. Training times dropped from hours to 30 minutes for less than €5 (about $5.70) per round, AWS I mentioned. The same infrastructure is now provided 60% faster time to market For new connected vehicle features and a 20% reduction in infrastructure costs.
When BMW migrated legacy systems with AI-powered tools, test creation time decreased from days to hours, resulting in time savings of more than 75%, while increasing test coverage by 60%, AWS I mentioned.
BMW i Ventures invests $300 million in building companies for physical AI
BMW i projects Fired Its third fund raised $300 million in April, bringing total capital under management to $1.1 billion. The third fund targets physical AI, agent AI, industrial software, advanced manufacturing technologies and materials. The fund invests from inception through Series B across North America and Europe. The goal is to back companies early enough to shape how the technology develops, not to buy in after the market has already formed.
European Central Bank Found This week, only 7% of eurozone companies reported extensive use of AI, although 70% say they use it in some capacity. BMW falls into a different category. Its US factory supports more than 12,000 jobs, its connected vehicle platform processes 16.6 billion orders daily, and its investment arm has committed $300 million to the layer of infrastructure beneath it.
