Ants, AI, and the Endless Immensity of the Sea
Let's explore what innovation has meant over time and what it can mean for you and your company right now and into the future.
We throw around the word innovation like it's a magic spell. Every strategy deck invokes it. Every tech keynote promises it. It’s the word you paste into the mission statement when you’ve run out of verbs. But before it became a buzzword, innovation had teeth. It represented risk, rebellion, sometimes a career-ending act of disobedience.
Today, we invoke innovation to signal progress, especially in fields driven by software, data, and now artificial intelligence. We build APIs to extend capability. We train AIs to detect patterns and generate content. But despite their high-tech wrapping, these tools only matter if they enable something new; something better.
Updating your toolkit is not innovation.
Let's explore what innovation has meant over time and what it can mean for you and your company right now and into the future. We’ll begin with a bit of forgotten history, where being an "innovator" could lead to imprisonment or worse. Then we’ll visit Schumpeter’s idea of creative destruction: the theory that growth depends on tearing down the old. From there, we’ll look at how innovation actually operates inside teams with lessons from ants, Netflix, and startups alike. We’ll close with four conditions that support sustainable innovation, and one final ingredient that can’t be easily measured.
Innovation as heresy
In 17th-century England, to be called an "innovator" was not a compliment. It was an accusation.
Henry Burton, a Puritan theologian, learned this the hard way. In 1637, he publicly criticized the Church of England and King Charles I for altering church rituals. His argument? These changes served the king's ambitions more than the faith. And Burton called these suspect alterations "innovations."
For this act, Burton was hauled before the infamous Star Chamber where he was charged with sedition, imprisoned in England's notorious Fleet Prison, and, just to underscore the point, had his ears cut off. Yes, really. Mutilated for challenging the status quo.
Ironically, King Charles would eventually lose more than his ears. In 1649, he was convicted of treason and beheaded.
The lesson isn't that innovation is bad. It's that innovation is dangerous; especially to systems that benefit from stasis. Throughout history, established powers have been deeply allergic to change. Today, we don't imprison people for questioning the way things work but we often ignore them, sideline them, or bury them in bureaucracy.
And a pervasive climate of innovation-by-slogan doesn’t help. Many organizations say they want disruption, but operate like factories of conformity. They reward consistency, punish deviation, and elevate blind execution over bold exploration. In an age when AI systems can mimic consensus at scale, the pressure to follow along has never been greater. It's easier to echo than to challenge.
Innovation as creative destruction
Fast forward a few centuries and Joseph Schumpeter, the Austrian economist, gives innovation a new identity. To him, it’s not a fringe act. It’s the core mechanism of capitalism itself.
Schumpeter called it creative destruction. Innovation, he argued, doesn't just improve the old, it eliminates it. It tears down entire industries, institutions, and assumptions to make room for something new.
This idea resonated through the 20th century and found new life in the startup era's favorite mantra: move fast and break things. It was thrilling, disruptive, and deeply flawed. Because while breaking things is easy, building better things is not.
We've seen what happens when innovation is unmoored from responsibility: social networks designed for growth but blind to consequence, algorithms that optimize for engagement rather than truth, AI systems that generate answers without understanding. Creative destruction, when practiced without regard for what comes next, becomes just plain destruction.
If Burton’s story was about innovation as personal risk, Schumpeter’s is about innovation as systemic upheaval.
Innovation as chaos
In a 2015 Harvard Business Review study, researchers Ron Ashkenas and Markus Spiegel observed that the most innovative teams didn't act like traditional organizations. They didn’t rely on rigid schedules, stage-gated approvals, or exhaustive documentation. Instead, they behaved more like adaptive ecosystems that were responsive, decentralized, and often improvisational. In other words, they acted like ants.
Innovation is like ants.
Stanford biologist Deborah Gordon has spent more than three decades studying how ant colonies operate, and her findings challenge everything we think we know about leadership and structure. In ant societies, there is no boss. No manager ant tells the others what to do. Yet the colony builds, adapts, defends, and survives.
Ants operate on local information. One ant bumps into another, senses a signal, changes direction. What looks like chaos from the outside is, in fact, a highly tuned system of dynamic response. No central planner. Just interaction, feedback, and emergence. Gordon writes, "A functioning organization with no one in charge is so unlike the way humans operate as to be virtually inconceivable." And yet, it works.
This is a helpful way to understand innovation. It rarely proceeds from a master plan. More often, it comes from dozens of small moves, local decisions, side-projects, and chance collisions. Innovation, like an ant colony, is shaped by proximity, timing, and responsiveness to change; not by formal reporting lines or top-down strategies.
But the ant metaphor likely rattles our instincts. It reveals a truth we often resist: innovation doesn’t scale through control. It scales through culture. You don’t get this kind of emergent behavior by enforcing compliance. You get it by building the conditions for adaptive response: autonomy, trust, visibility, and feedback.
Innovation is chaotic, unpredictable, and messy.
Four conditions for a culture of innovation
So what, if anything, can you do now to improve your culture of innovation, to foster more creative thinking, and protect innovators within the organization? It's not about adding another "innovation lab" or rebranding your teams in some way. It's about changing the environment for everyone in the company.
The good news is you don’t need a complete reorg. There are tangible, human-scale changes that make a real difference. Here are four conditions that, while they won’t guarantee innovation, will make it far more likely to happen. And more likely to last when it does.
1. A central mission with a loose structure
Ted Nelson, the person who coined the words "HyperText" and "Hypermedia" in a 1965 paper, once said, "Anyone, anywhere should be able to write anything about everything without having to ask for permission." He was talking about the World Wide Web, but the spirit of that declaration applies just as much to today's fast-moving AI ecosystems and to the organizations navigating them.
The most innovative teams aren't those with the most rules. They're the ones with the clearest sense of why they exist and the fewest constraints on how they get there. Clarity of purpose, paired with structural looseness, allows for "uncontrolled" exploration without destructive chaos.
Amazon famously lets teams spin up services without central approval, bounded only by operational feasibility and shared infrastructure. OpenAI invites developers worldwide to remix, fine-tune, and extend its models; effectively open-sourcing the future of AI interaction. Netflix avoids micromanagement by emphasizing "context, not control" and providing high-level guidance to teams while trusting teams to make tactical decisions.
The thread running through these examples is autonomy with alignment. When people understand the mission, they don't need micromanagement. They need room to move. Innovation thrives not where the rules are strictest, but where the mission is clearest.
2. Maximize learning through open sharing
Innovation isn't about being right. It's about learning faster.
Peter Senge, in The Fifth Discipline, offered a quietly radical view of organizational life. He wrote, "The only sustainable competitive advantage is an organization's ability to learn faster than the competition." Not to ship faster. Not to hire smarter. But to learn; continuously, collectively, and out loud.
This idea is more relevant than ever. In the age of AI, where models improve through exposure and feedback, learning isn't just a human virtue; it's an architectural principle. The best teams don’t just move fast. They reflect quickly. They debug, document, and share what didn’t work.
And in software more broadly, tools like continuous integration and rapid deployment have made learning an infrastructure-level feature. They reduce the cost of error and increase the tempo of feedback. Every commit, every test, every release is a potential lesson.
To innovate effectively, your organization needs to think like a learning system. Not just a production pipeline. That means encouraging reflection, making failure visible, and treating each mistake not as a deviation, but as a data point. Because the faster you learn, the faster you grow.
And the good news is this ability to learn-by-doing is not a skill today's AI systems can mimic.
3. Constant experimentation
Innovation doesn’t emerge from endless planning sessions. It comes from trying things that might not work—and learning from the results.
In The Chaos Imperative, Ori Brafman makes a compelling case for why chaos isn’t something to eliminate, it’s something to harness. Brafman describes how injecting a little disorder into rigid systems can yield surprising breakthroughs. Unstructured time, chance encounters, even deliberately random constraints can open up possibilities that structured thinking can’t reach.
This principle is especially powerful in software development and AI research. APIs and models are inherently modular, reproducible, and disposable. You can build one in a day, test it in an hour, and throw it away in a minute. This flexibility makes them ideal substrates for experimentation. You don’t need permission to prototype. You just need curiosity, tools, and a culture that values the attempt even if it fails.
The lower the cost of failure, the more experiments you can afford. And the more experiments you run, the better your odds of hitting on something meaningful. Not every test leads to a breakthrough, but each one teaches you something about the user, the system, or the limits of your assumptions.
In a world where technology moves fast and conditions change daily, a culture of constant experimentation isn’t just a nice-to-have. It’s your best hedge against irrelevance.
4. Freedom to explore the next horizon
Most organizations ride one wave of innovation and then try to hold their balance as long as possible. The truly great ones are already paddling toward the next.
In Jumping the S-Curve, Paul Nunes and Tim Breene describe a pattern seen in high-performing companies: they anticipate the plateau. Instead of waiting for growth to stagnate or disruption to arrive, they invest early in the next wave; often while the current one still looks strong. This leap requires more than vision. It demands institutional permission to explore what doesn't yet seem useful.
Technological progress tends to follow S-curves: slow beginnings, a phase of rapid acceleration, then a flattening as the approach matures. Legacy systems get optimized. Processes get standardized. And then someone else shows up with a new curve—one that starts below yours, but grows faster.
In the software world, this happens constantly. APIs opened up systems that were once closed. Now, generative AI is starting to upend how we create, design, and communicate. These early curves look messy, inefficient, even trivial. But they are the proving grounds for what’s coming next.
Smart organizations create protected space for teams to explore those early curves. Google’s “20% time,” Amazon’s “two-pizza teams,” and OpenAI’s community sandbox model all represent different ways of granting that freedom. The point isn’t to guarantee immediate returns. It’s to build capacity for what's beyond the horizon.
Exploration is a long-term bet on relevance. And while not every experiment pays off, failing to explore is the surest way to fall behind. If the current S-curve defines your success, the next one will determine your survival.
A culture that makes space
These four practices, a clear mission, shared learning, constant experimentation, and future-minded exploration, aren’t checkboxes. They’re persistent conditions. Together, they can create a safe, supportive place where people can think creatively about the problems that face the company today and the challenges that lie ahead.
This is the soil in which innovation grows. It protects the rebels, honors the learners, invites the explorers, and trusts the swarm. When you build this kind of environment, you don't need to chase innovation. You just need to listen for where it's already trying to happen.
Don’t forget inspiration
Innovation doesn’t just need process. It needs inspiration. That’s what makes people take risks, stay curious, and push past setbacks. It’s the difference between compliance and commitment.
Netflix reminds its leaders of a quote from Antoine de Saint-Exupéry, the French pilot and author of The Little Prince:
"If you want to build a ship, don't drum up people to collect wood and don't assign them tasks and work, but rather teach them to long for the endless immensity of the sea."
That longing, that sense of possibility, is what fuels genuine innovation. It’s why the best leaders don’t just assign roadmaps. They cultivate vision. They give people something worth striving for, something just out of reach.
Too often, innovation programs focus on inputs: budget, talent, OKRs. But it’s the intangible sense of wonder, the thrill of discovery, the permission to imagine something that carries people through the hard parts.
So yes, APIs and AI models can be our tools. Processes can be our guardrails. But the real innovation happens when someone longs for the sea.
That’s why the "I" in API, and increasingly in AI, still stands for innovation.