Everyone's talking about Artificial Intelligence as the next big thing, often comparing this moment to the dot-com boom of the 1990s. It feels like we're on the cusp of a revolution that will change everything, just like the internet did. But what if we're looking at the wrong decade for comparison? The truth is, AI today has more in common with the state of computing in the 1960s—powerful, promising, but locked away and accessible only to a select few.
The Three Walls Holding AI Back
Before AI can become the new internet, we have to tear down three massive walls: crushing costs, extreme concentration of power, and a sheer lack of capacity.
Imagine wanting to build a website in the late '90s, but instead of a cheap hosting plan, you had to rent a government-sized supercomputer. That's essentially the situation for many AI developers and startups today. The U.S. AI Action Plan itself points out that companies often need to sign long-term, prohibitively expensive contracts with a handful of giant cloud providers, or 'hyperscalers'. This concentration of power not only stifles competition but also creates artificial barriers, like high fees for moving your own data, that keep smaller players out of the game.
A History Lesson from the PC Revolution
To understand the path forward, we need to look back further than the internet—to the transition from room-sized mainframes to the personal computer. In the 1940s and '50s, computing power was so rare and expensive that only governments and major universities could afford it. My smartphone today is vastly more powerful than those early machines.
So, what changed? The revolution was kickstarted by companies like IBM, which commoditized computer hardware, and Microsoft, which created a standardized operating system. This wasn't about one magical invention; it was about creating open standards that allowed anyone to build, innovate, and connect. The internet followed a similar path with the TCP/IP standard, a common language that let any computer talk to any other, breaking down the walled gardens of services like AOL and unleashing a wave of creativity.
AI is currently stuck in that pre-PC, mainframe era. We need our own 'IBM moment'—a shift towards open standards and an accessible marketplace for AI resources.
The Path to a Truly Open AI Future
This transformation won't happen by government decree alone, but it does require smart guidance. The goal is to create market forces that naturally drive down costs and spur innovation. When companies can easily compare AI computing options and switch between providers without massive penalties, true competition begins. When a developer can access the tools they need without months of planning and a bottomless budget, innovation accelerates.
This isn't just a domestic issue; it's a global one. Nations like China are already implementing national strategies, creating state-run exchanges for computing power and subsidizing their AI champions. For Western values to lead the AI-driven future, we need a strategy that promotes an open, interoperable ecosystem. By setting standards that our allies and even adversaries can adopt, we build influence and strengthen the entire democratic technology stack.
Governments can play a crucial role by ensuring markets remain competitive, prioritizing smaller enterprises in procurement, and even creating strategic reserves of computing power for crises, much like the strategic oil reserve.
Key Takeaways
The journey to AI's full potential is just beginning. To get there, we need to remember the lessons of the past and focus on what truly drives technological revolutions.
- The Right Analogy: AI is currently more like 1960s computing than the 1990s internet, limited by cost and access.
- Major Hurdles: The biggest barriers are high costs, concentration of power in a few companies, and insufficient infrastructure.
- The Power of Standards: Open standards, like those that sparked the PC and internet revolutions, are needed to democratize AI.
- Market-Driven Innovation: The solution lies in creating a competitive market where innovation can flourish, guided by government policy that prevents monopolies.
- A Global Race: The development of these standards is a global competition, and establishing an open, accessible model is key to leading the way.