The Great Value Inversion: Earth over Algorithms

My background connects what most see as separate worlds: designing digital experiences and understanding the materials that make them possible. From this system perspective, I see a transformation that challenges the foundation of our knowledge economy. The current rise in artificial intelligence, specifically LLM models, has begun the transition that makes our most distinctly human capability—intelligence—economically worthless,  whiles the raw materials that power this revolution are becoming more precious than ever.

We are witnessing an overproduction of intelligence, humanity’s defining trait, which is becoming economically cheap, while raw materials from the earth gain unprecedented strategic value.

AI is not just disrupting jobs; it is inverting the entire relationship between the immaterial work we create and the material world that sustains it.

Cognitive Labor is Becoming Worthless

Tasks that once required years of specialized training—writing code, analyzing data, creating digital content—can now be automated in seconds. As someone who has spent years helping companies build digital experiences and product innovations, I’m watching this transformation devastate entire industries. The routine analysis and content generation that built our knowledge economy are being hit hardest. Software engineers, data analysts, content creators—roles that commanded premium salaries—face an uncomfortable truth: their cognitive output is becoming reproducible at scale.

This isn’t limited to entry-level work. Senior designers creating user interfaces, product managers crafting digital strategy, even creative directors conceptualizing brand experiences are finding AI can produce similar outputs at near-zero marginal cost. The immaterial aspects of intelligence that built entire knowledge economies are rapidly becoming commoditized. Not because AI truly “thinks,” but because it doesn’t need to for most commercial applications.

The irony is stark: as our digital creations multiply exponentially, their individual economic value plummets toward zero.

Second: The Immaterial Lives in Material Bodies

Here’s what we’ve missed while celebrating the “immaterial magic” of AI, a critical detail has been overlooked: every algorithm lives in hardware.

As Kate Crawford demonstrates in “Atlas of AI,” artificial intelligence is neither artificial nor immaterial—it depends entirely on physical materials extracted from the earth.

Crawford’s groundbreaking research reveals how AI systems require vast planetary resources: chips require rare earth elements, batteries need lithium, servers demand copper, and manufacturing depends on geopolitical stability.

Every machine learning model, every smartphone displaying our digital designs, every server farm processing our data depends on materials extracted from specific geographic locations. Crawford traces AI’s supply chains from lithium mines in Nevada to rare earth processing in China, showing how our digital dreams are built on very physical foundations. The devices we design for—the platforms that host our immaterial work—cannot exist without this material substrate. Yet while we’ve been mesmerized by AI’s cognitive capabilities, the real economic value has quietly shifted to the material foundation that makes it all possible.

Raw Materials Boom on the Rise

As machine learning makes intelligence abundant, geography becomes destiny again.

The Democratic Republic of Congo doesn’t just supply cobalt; it supplies critical infrastructure for our entire digital civilization. Countries sitting on lithium deposits, cobalt mines, and rare earth reserves are no longer just resource providers—they hold the keys to the digital kingdom.

This scarcity is accelerating. While AI can generate infinite content, the materials needed to run AI systems are finite and geographically concentrated. Current recycling efforts recover only 17% of electronic waste globally, far below the scale needed to meet growing demand. According to a UN report, humanity produces 62 million tonnes of electronic waste every year, of which less than a quarter is formally recycled, leaving valuable and finite resources in landfills.

Unlike software, you can’t simply copy and paste lithium or manufacture rare earth elements from thin air.

The strategic implications are profound. Access to stable supply chains for critical minerals now matters as much as, if not more than, access to the best programmers.

Critics rightfully point out several limitations to this analysis. Not all human intelligence can be automated—creativity, emotional intelligence, and complex problem-solving remain irreplaceable and may become more valuable as AI handles routine tasks. Additionally, recycling technologies and material substitutes are advancing rapidly, potentially alleviating scarcity pressures. For example, the IEA has found that while e-waste is at a record high, the production of recycled battery metals is growing rapidly and could reduce the need for new mining by 25% to 40% by 2050.  The timeline may be longer than anticipated, with regional variations and policy interventions capable of redirecting these trends entirely. Historical precedent suggests markets adapt to resource constraints through innovation and price signals, just as past predictions of material shortages proved premature. Yet these valid concerns don’t negate the fundamental pattern: certain cognitive work is becoming cheaper to produce while physical infrastructure becomes more strategically important. Even if the changes unfold over decades rather than years, organizations ignoring this shift risk being caught unprepared.

The Path Forward

I often work with companies for whom the conversation ends when features ship; they don’t clearly articulate the materials footprint of their choices, whether they’re sustainable, or what they learned about the physical systems enabling their digital dreams.

The future belongs not to those who can think like machines, but to those who understand that our digital dreams are built on physical foundations.

Organizations that thrive will combine algorithmic capabilities with sustainable materials strategies to navigate the complex relationships between human creativity, platform value, and material access.

As Crawford reminds us, artificial intelligence is a technology of extraction—from energy and minerals to human labor and data. Understanding this material reality is essential for building technologies that serve humanity rather than depleting our shared planetary resources. The companies that succeed won’t just have the best algorithms or the most materials—they’ll understand how these elements work together sustainably.