When we try to predict the future of agriculture, we obsess over obvious questions.

Will AI transform farming?

How will climate change affect yields? 

But attempting to make predictions like this often blinds us to something more fundamental: the hidden assumptions shaping how we think. These assumptions, when wrong, can destroy billions in value overnight or leave critical systems vulnerable to collapse.

The Trillion Dollar Day

On January 27th 2025, something remarkable happened in the world of technology. NVIDIA, a company that had become synonymous with artificial intelligence, lost over a trillion dollars in market value. That's not just a big number - it was the largest single-day drop in market value in corporate history. 

The company hadn't failed. There was no scandal. Their earnings handily exceeded expectations. But in a matter of hours, the market fundamentally questioned its assumptions about the future of AI infrastructure, and about what mattered most in the race to domination of the AI field.

The NVIDIA story wasn't really about one company's stock price. It was about how our assumptions shape our view of the future. The market had built its narrative on a seemingly solid foundation: AI development requires massive computational power, which means more specialized chips, in bigger data centers. And all of this meant that NVIDIA's dominance would continue to grow. 

But when that assumption wavered, when DeekSeek showed seemingly impossible results from such minimal GPU capacity, it provoked a question  around whether AI's future really demanded ever-expanding GPU infrastructure - and the NVIDIA dominance narrative collapsed. 

A trillion dollars of value vanished because an assumption about the future proved less certain than we thought.

Twenty-Two Ships and The Internet's Backbone

While all this was making headlines, another story was unfolding literally beneath the waves of our oceans. Between November 2024 and January 2025, a series of incidents damaged multiple undersea cables - the physical infrastructure that carries 97% of global internet traffic. These weren't just random accidents. They represented a systematic, and probably deliberate, exposure of one of our modern economy's most critical vulnerabilities. 

And to make things worse, this incident revealed that we only have twenty-two ships in the entire world that are capable of repairing these vital undersea cables. Twenty-two ships to maintain the physical infrastructure that our global digital economy depends on. When multiple cables were damaged simultaneously, this limited repair capacity became a critical constraint. 

Unlike NVIDIA's dramatic stock drop, this story received relatively little attention, despite potentially far greater commercial impact. But it's a perfect example of how focusing on one part of a system - the cables themselves - can blind us to crucial vulnerabilities in another part - our ability to repair them.

What Are We Not Seeing?

In agriculture, we face similar challenges. When we simplify complex systems - whether it's soil health, water management, or supply chains - we risk overlooking critical dependencies. Just as the market overlooked repair capacity in undersea cables, we might be overlooking crucial connections in our agricultural systems. We might be failing to question hidden assumptions that are shaping our view of agriculture's future, and potentially making us vulnerable to hidden threats.

Consider just a few assumptions quietly shaping agriculture's future:

• That water rights and access will remain relatively stable
• That international supply chains will continue functioning smoothly
• That consumer food preferences will evolve gradually rather than shift suddenly
• That farm labor shortages will eventually stabilize

Each of these assumptions, if wrong, could upend entire agricultural business models. Yet how often do we explicitly examine them?

Tools for Testing Assumptions

In our investing and advisory work, we rely heavily on key frameworks and tools to challenge and expand our thinking. If  you want to think more systematically and test your assumptions, we can highly recommend looking further into:

• The Wide Lens by Ron Adner encourages us to view systems in terms of complex interconnections of participants and deeply understand alignment of incentives.
• The premortem flips traditional storytelling by having teams imagine a project's failure, then trace its hypothetical path backward. This creative technique harnesses narrative imagination to uncover potential risks.
• In 2024, we put these approaches to work in a report called Navigating a Future of Cross-Sectoral Forces in collaboration with AgriFutures Australia.

The future of food and agriculture won't be determined by how accurately we predict specific changes. It will be determined by how wisely we identify and challenge our hidden assumptions today. The tools and frameworks highlighted above aren't just helpful extras—they're essential defenses against the trillion-dollar blindspots that lurk in every industry's future.

This article is drawn from a session we co-developed for evokeAG 2025 in Brisbane, Australia.

‍