Industrial AI strategy and competitiveness

• Industrial AI differs from consumer AI: it uses data most models have never touched and operates where hallucination is catastrophic
• Europe's edge is data, not capital: the US invested $286 billion in AI last year against $21 billion in Europe
• Siemens is investing more than €1 billion in industrial AI over three years
• Industrial AI is positioned as central to Europe's digital sovereignty and competitiveness
• Three shifts reshape manufacturing: coordinated AI, self-adapting automation, and conversational engineering tools
• Customers now ask for entire operations, with Siemens citing scrap cut by up to 20 percent at P&G
• 59 percent of UAE manufacturers expect transformational AI adoption by 2030, against 42.2 percent globally
• Schneider Electric won the Mining Magazine AI in Action Award for measurable industrial AI impact

‍

Data, context, and trust as the foundation for AI

• The shift is from collecting data to contextualizing it: factories face a meaning problem, not a data problem
• AI on SAP Digital Manufacturing data tied temperature violations to a mixing operation and one routing step, pending expert validation
• A semantic thread across the V-model and a knowledge graph are proposed as the basis for traceable engineering AI
• Letting AI agents edit a bill of materials unsupervised carries serious operational risk
• Scaling industrial AI requires clear decision rights and verified outcomes
• A federated AI platform lets semiconductor companies train shared models while exchanging only masked model updates
• New research combines physics-informed features with machine learning for tool wear prediction

‍

Agentic and physical AI

• Siemens ENVOY runs as a shared foundation across hundreds of product teams and stood up nine agents in one week with AWS
• AWS launched an open Physical AI Toolchain with NVIDIA to close gaps between lab robots and edge deployment
• AWS research across 4,420 startups found one in seven building physical AI and only 26 percent saying their economy is ready
• Embedding AI in sensors addresses latency, power, and autonomy limits
• Edge AI is framed as essential for data sovereignty as cybersecurity risk rises
• Siemens Industrial AI Suite targets the 6 to 12 months it can take to integrate one AI model into a machine
• Autonomous Regulated Manufacturing applies closed loop agentic AI across life sciences handoffs

‍

Digital twins and digital threads

• Mandelli validates machining with the Siemens CNC Digital Twin before production
• GROB uses a continuous Digital Twin to speed commissioning and reduce engineering effort
• BSH reports up to 50 percent faster time to market with a digital thread on Siemens Xcelerator
• HD Hyundai and Siemens opened a Global Marine Center in Busan to build an AI powered Digital Thread
• Siemens Run My Virtual Machine brings the CNC experience to the desktop
• Many digital twin needs are simpler than assumed and should start with basic visibility

‍

MES, cloud, and IT/OT convergence

• MES is moving from a system of record to a system of intelligence
• Decospan cut custom coding in its SAP Cloud MES rollout using the delaware DMforMill package
• Siemens Opcenter connects ERP, warehouse, and production systems for end to end traceability
• Siemens added the Paessler PRTG probe to its Industrial Edge Marketplace to close IT/OT visibility gaps
• ESAB is progressing toward go live on RISE with SAP S/4HANA Cloud Private Edition
• KSB scaled usage based revenue for KSB Guard with SAP Subscription Billing and Sopra Steria

‍

Deals, partnerships, and events

• Schneider Electric agreed to acquire PTC in an all cash transaction valued at $22.6 billion
• Bosch and Microsoft are combining manufacturing and AI expertise for greater customer value
• Schneider Electric and Automatique & Industrie built a cybersecure automation solution for renewable energy sites
• IMTS 2026 in Chicago featured the launch of Meet at the Machine with TRAK Machine Tools
• SEMICON West 2026 featured a talk on federated AI for semiconductors
• CAMX in Atlanta highlighted AI, automation, and inspection in composites

‍

Thanks to Anshul Sahu, Roland Busch, Peter Koerte, Boris Shalumov, Silvio Saouaf, Christoph Golinski, Helenio Gilabert, Ravi Kunju, Rocio Thompson, Sascha Schlick, Dennis Dombrowski, Ricardo Marroni, Cedrik Neike, Ujwal U Ramdurg, VÍCTOR Martínez, Neil Barua, David Randle, Tanja Rueckert, Edoardo Martini, Holger Wild, Clairy Chow, Sameet Saraf, Jean-Baptiste FUNTEN, Shafeen Nayan Veettil Muhammed, Mustafa Savaşcı, Semih Cemil Dumlu, Tobias Hesse, Sri Elaprolu, Chris Stevens, Jerome Tourdiat, Harsh Parmar, Stefan Weisenberger, Alex Teo, Stefan Köhler, Alexander Stukenkemper, Kurt Höller, Michael Taesch, Ananth Kommuri and Shival T., and everyone else who contributed insights to this edition.

Find the full list of posts and voices on LinkedIn: https://www.linkedin.com/pulse/best-linkedin-cw-41-42-smart-manufacturing-industrial-thomas-allgeyer-bxyme/