Davos 2026: The Fusion of AI and Manufacturing
I've been working with manufacturing operations across Europe for years, watching factory leaders navigate the gap between AI hype and operational reality. At Davos 2026, that gap collapsed in real time.
This year wasn't about debating whether AI matters. World leaders, military generals, and technology builders were talking about power grids, Arctic security, and who controls manufacturing data. The conversation moved from "if" to "how fast", and if you're deciding today which systems will run your factory over the next decade, the window to lead is closing.
Here's what I understood from Davos: The global system is shifting from optimisation to acceleration. And manufacturing is becoming a strategic asset again.
What I'm sharing below isn't a Davos report. It's a translation of what these shifts mean for the decisions you're making right now about factory systems, brownfield modernisation, and how to retain knowledge as your best operators retire.
New Times Are Coming: Defence as a Prerequisite for Economic Transformation
At Davos, beneath the political theatre around Greenland and Arctic security, a fundamental shift became clear: Europe's ambitious economic agenda cannot succeed without first solving its defence gap.
Russia dominates the Arctic militarily, nuclear assets, icebreakers, missile-carrying submarines. Denmark cannot secure strategic locations like Greenland or Bornholm alone. Critical infrastructure from missile installations to Norwegian gas routes across the Baltic Sea requires collective protection.
Canadian Prime Minister Mark Carney framed it bluntly: "Our old, comfortable assumption that our geography and alliance memberships automatically conferred prosperity and security is no longer valid."
NATO is recalibrating. The United States expects European partners to increase defence spending substantially and take responsibility for their own security. Despite different national appetites for military engagement, Europe understood one thing at Davos: defence investment is no longer optional if industrial transformation is the goal.
Why this matters for manufacturing:
Supply chains, energy systems, data infrastructure, and industrial capacity are now strategic assets. Factories are part of national resilience. In this reality, military force, not international law, protects these assets.
Manufacturing capacity has become inseparable from national sovereignty. If your factory systems cannot adapt to this new reality quickly, you're not just operationally slower, you're strategically exposed.
Europe is waking up. And that changes the context for everything that follows, from infrastructure investment to AI strategy.
Germany and Japan: AI and Infrastructure as National Strategy
Two speeches at Davos stood out. Not for rhetoric, but for budget commitments and strategic clarity.
German Chancellor Friedrich Merz and Japanese Finance Minister Satsuki Katayama outlined remarkably similar visions. Both matter directly to our business, and to our customers' operations.
Germany: From Nostalgia to Industrial Power
Merz delivered one of the most consequential addresses at Davos, declaring the old world order broken and calling on Europe to act with power, unity, and realism.
On defence, he was blunt: "We must invest massively in our ability to defend ourselves, and we are doing this." Germany's defence spending will rise to 5% of GDP, what Merz called "a huge increase" designed to "reduce our economic and technological dependencies."
But it was his focus on infrastructure and AI that revealed the deeper strategy, and what I see as the critical opportunity for manufacturing.
"In the coming years, you will see massive investments in state-of-the-art power plants, power lines, and heat supply," Merz announced, citing a €500 billion commitment to rebuild Germany's physical backbone.
Then came the part that matters most for manufacturing:
"At the heart of our efforts lies digital transformation. Artificial intelligence requires an industrial scale. Germany has one of the world's largest pools of industrial data. That is just one reason why we are investing in high-performance AI gigafactories, speeding up the expansion of data centres, and creating the digital infrastructure for a competitive AI economy in Germany.
Our research and technology policy is guided by a new high-tech agenda. We are global leaders in many areas of cutting-edge research. We want to ensure that innovation gets to market more consistently, building industries of the future.
And whoever wants to invest in the future—let me be clear—you will find a very strong partner in Germany. We want to be a leading investment location for global capital. Our policy is to mobilise private sector investments in infrastructure, high-tech, and industrial transformation—with clear rules, strong institutions, and long-term reliability."
Japan: Automation as National Survival
Japan is moving on a parallel track, but with even sharper urgency driven by demographic reality.
The government is steering over $330 billion USD in public-private investments into AI and semiconductor sectors, supported by targeted tax reforms to boost domestic R&D. As Finance Minister Katayama explained, Japan is "shifting from a deflationary cost-cutting economy to a dynamic growth-oriented one driven by bold investment and productivity gains."
Here's what makes Japan's approach different, and what I think European manufacturers should pay attention to: they're not debating whether automation displaces jobs. Facing a shrinking and ageing workforce, they've embraced it as essential infrastructure.
"Everybody is happy to get robotics and AI into the real world," said NEC Chairman Nobuhiro Endo. Japan's societal openness to "physical AI" and job-specific automation gives it a unique competitive edge. There's no political resistance, just pragmatic implementation.
Geopolitically, Japan is also recalibrating. "We have been between the United States and China for long years… we know both of them very well," Katayama remarked. Japan is deepening alignment with Indo-Pacific allies to safeguard access to critical technologies and materials while maintaining strategic balance.
What Germany and Japan Share, and Why It Matters
These aren't random policy announcements. They're coordinated national strategies from two of the world's most significant manufacturing economies. Economies, where decisions directly impact our business and our customers.
Three critical similarities:
Both have diagnosed the geopolitical shift and are reinforcing defence capacity as a prerequisite for economic security
Both have identified their challenges and committed massive capital, Germany €500B+ in infrastructure, Japan $330B+ in AI and semiconductors
Both are focusing on AI and infrastructure to reinforce their strong manufacturing traditions, not replace them
Germany sits on decades of industrial process data. Japan has unmatched precision manufacturing expertise and social acceptance of automation. Both are now mobilising this heritage as strategic fuel for the AI economy.
Here's what I'm seeing in factories and what this means for manufacturing leaders:
We should expect massive acceleration of new technologies being introduced and implemented at a large scale in the coming years, not as pilot projects, but as national infrastructure.
The window to define your individual strategy is right now.
Because when governments align defence spending, energy infrastructure, and AI investment into a single framework, factory systems are no longer just operational decisions. They're strategic assets in a much larger transformation.
The assumption that's breaking: The idea that you can wait 3-5 years to "see how AI develops" before committing to new architecture. That timeline just compressed to 18 months, and the cost of retrofitting later means learning on everyone else's timeline.
Jensen Huang: How AI Capabilities Actually Get Built at Scale
Germany and Japan are committing hundreds of billions to defence and AI infrastructure. But how do you actually build AI capabilities at an industrial scale?
Jensen Huang, CEO of NVIDIA, provided the most useful framework at Davos for understanding what's required.
He described AI as a five-layer industrial stack:
Energy — the foundation, because AI processes data in real time and needs power to generate intelligence
Chips & computing infrastructure — where NVIDIA plays, providing the hardware for AI computation
Cloud infrastructure — the backbone for training and deployment
AI models — where most people focus
Applications — where real economic value is created
Most debates obsess over layer four. Most value gets created at layer five.
This is where manufacturing becomes decisive.
Huang emphasised something critical: AI models are becoming domain-specific. They can now "speak the language" of proteins, polymers, chemistry, and physics. Not generic chatbots. Industrial intelligence.
He also talked about agentic AI, systems capable of performing complex, multi-step tasks, making decisions, and reasoning through problems. This is AI as a partner in productivity, not just a tool.
And then he made a direct appeal to Europe:
"This is your opportunity to leap past the era of software and fuse your industrial capability with artificial intelligence."
That word—fusion—is the key insight. And it's what I see as the defining opportunity for European manufacturing.
Europe doesn't need to win consumer platforms. It can win physical AI. The combination of strong manufacturing traditions, research capabilities, and AI creates something new. That's the opportunity.
Here's Why This Matters for Your Factory Tomorrow Morning
When governments are mobilising defence budgets to protect investments in AI-driven economic modernisation, factory leaders need to ask themselves one critical question:
Are you locked into legacy hardware and systems that can't run modern edge AI?
Because here's the reality I'm seeing: Generic AI tools won't understand your specific alloys, temperatures, or tolerance requirements. Domain-specific Industrial AI will.
Obviously, we are in this application domain, and this is where the real battle happens.
Huang also stressed something important: democratizing AI. Making it accessible broadly across society, supported by infrastructure investments. Because if the average factory, or the average manufacturer, is just watching AI from the sidelines, they're going to feel left out.
The question isn't whether we're in an AI bubble. The real question is: are we investing enough?
Elon Musk: From Abundance to Consciousness. Why Manufacturing Matters at Civilisation Scale
Jensen Huang gave us the five-layer framework for building AI at scale. Elon Musk showed us what happens when you actually execute on those layers, and why it matters beyond quarterly results.
The interview with Musk cut through a lot of noise. He doesn't frame AI as a feature or even as infrastructure. He frames it as a civilisation-scale system. His mission across SpaceX, Tesla, robotics, and AI is consistent: preserve and propagate human consciousness, even beyond Earth through redundancy and scale.
Super interesting stuff. But several points connect directly to what we've heard from governments and technology builders and to what I'm seeing in manufacturing operations.
Energy: The Real Bottleneck, and the Solution in Space
"The limiting factor for AI deployment is fundamentally electrical power."
This addresses Jensen's Layer 1. And Musk speaks clearly about the opportunities in solar and solutions in space. Considering his success with SpaceX and experience in reducing the cost of shipping equipment into orbit, solar-powered data centres in space can become sustainable and feasible solutions.
"Solar is by far the biggest source of energy. Even on Earth but certainly beyond Earth, the sun rounds up to 100% of all energy."
"It's a no-brainer for building solar-powered AI data centres in space. It's also very cold in space… 3° Kelvin… [so] it's just cooling. A very efficient cooling system. The lowest cost place to put AI will be space… within two years, maybe three at the latest."
AI doesn't float in the cloud. It runs on electrons and physics. And Musk is already solving for Layers 1, 2, and 3 of Jensen's framework. Not theoretically, but with launch schedules.
What this means for factory energy planning: If your facility can't handle 3x the compute load when you deploy real-time AI quality control, your AI strategy is already constrained by infrastructure you didn't budget for.
From Tesla to Robotics: What Data Accumulation Actually Enables
Another interesting insight is how Musk leverages the accumulation of data from one domain—Tesla electric vehicles—to a new domain: autonomous robotics.
"Tesla is obviously about sustainable technology… [and] sustainable abundance. The Tesla full self-driving software… we update it sometimes, once a week."
"Humanoid robotics will advance very quickly… we do have some of the Tesla Optimus robots doing simple tasks in the factory. By the end of next year… we'd be selling humanoid robots to the public… very high reliability, very high safety."
Here I read two critical lessons for manufacturing:
First: To take advantage of future AI applications, you have to prepare yourself and invest in the infrastructure to manage impactful data, not big data, but data that teaches systems how to perform in your specific domain.
Second: AI models based on general knowledge will become more narrow and domain-specific. This directly addresses Jensen's Layer 4. The value isn't in generic models, it's in models trained on your processes, your materials, your operational reality.
The pattern I'm seeing in factories: Tesla updates its AI weekly. Conventional manufacturing systems get upgraded once a year and you pay for it every time. That difference in learning velocity is the difference between leading and catching up.
Abundance Is Mechanical and It Solves Real Problems
Musk's vision of abundance connects directly to what we heard from Japan about demographic reality:
"With robotics and AI… this is really the path to abundance for all. The only way to… give everyone a very high standard of living… is AI and robotics."
"If you have… ubiquitous AI… and ubiquitous robotics… You will have an explosion in the global economy. There'll be more robots than people."
This isn't science fiction. It's a response to ageing societies and labour shortages. Japan understands this. Germany is moving toward it. And there's symmetry here with what Jensen said about differentiating between the purpose of human work and the task itself.
At Jensen's Layer 5—the application level—the opportunity for manufacturing starts with industrial AI applications that first understand what the purpose is and what tasks are to be automated before moving toward deeper automation.
Consider what this means on a factory floor: A single robot learning to perform quality inspections doesn't just replace one worker, it creates a replicable capability. Train one, deploy a thousand. That's not automation. That's institutional memory at scale.
Vision Drives Transformation
What makes Musk's approach relevant isn't just the technology, it's what drives it: deep curiosity about fundamental questions and a crystal-clear vision of where it all leads.
His SpaceX mission is a perfect example: extend life and consciousness beyond Earth to ensure that "the light of consciousness is not extinguished." He sees humanity as a tiny candle in vast darkness, fragile, but worth protecting through redundancy and scale.
This clarity of purpose, understanding not just what to build, but why it matters, is what enables real transformation. It's what turns incremental improvement into paradigm shifts.
Manufacturing needs this kind of vision. Not vague statements about "digital transformation," but clear answers to: What are we actually trying to preserve and extend? What bottlenecks are we solving? What does abundance look like in our specific context?
With that clarity, pilots and experiments become learning steps toward a destination. Without it, they remain isolated projects that never scale.
Musk closed with something memorable. He said he'd like to die on Mars "but not on impact," which got a laugh. And then this:
"It's better to be an optimist and be wrong than to be a pessimist and be right."
That's not just a clever line. It's a stance. And it's the stance manufacturing leaders need right now, not blind optimism, but the courage to build for a future worth living in.
What This Means for Factory Decision-Makers
We've heard from governments realigning defence and economic policy. We've heard the framework for building AI at scale. We've heard the vision of what abundance could mean.
Now the question is: what does this mean for the decisions you're making this quarter?
Let me make this concrete.
Here's what Davos 2026 quietly confirmed:
AI is becoming a strategic infrastructure, like energy or transport
Manufacturing data is a national and corporate asset
Automation and robotics are responses to demographic reality, not just efficiency plays
Energy capacity and system architecture matter more than dashboards
The cost of inaction is rising exponentially, retrofitting later means learning on everyone else's timeline
If you're a manufacturing leader making technology decisions today, here's the critical choice:
Is it worth investing in outdated technologies where AI is just an add-on? Or should you focus on AI-native solutions?
Two questions can help guide your answer, and your next AI experiments and focused pilots:
How do we modernise brownfield factories?
How do we retain knowledge as experienced operators retire?
These aren't abstract strategy questions. They're urgent operational realities that AI-native systems can address in ways legacy platforms with AI bolt-ons simply cannot.
The question is no longer "Should we explore AI?"
The real question is:
Are your factory systems designed for acceleration or just observation?
Because the world is accelerating. Governments know it. Technology leaders know it.
The only real risk now is standing still while everyone else learns faster.
Closing Thought
Davos 2026 wasn't about optimism or fear. It was about alignment.
Energy, security, infrastructure, AI, and manufacturing are converging into a single system. The fusion of AI and manufacturing isn't coming in five years. It's happening now. Those who design for that reality will shape the next decade.
The rest will be forced to adapt later, at a much higher cost.
If you want to see what this looks like inside your factory, not on a Davos stage, let's map it. Three-month pilot. Your production data. Your real constraints. We'll show you where the acceleration opportunity sits in your operation with a real use case.
Because the leaders who move now won't just survive the next decade.
They'll define it.