There's no doubt we're in the middle of the Fourth Industrial Revolution. But if we zoom out, this moment looks less like a threat and more like the opening chapter of a massive opportunity. Every major technological revolution has followed a remarkably consistent pattern, and AI appears to be right on script.
Over the past 200 years, the sequence has been strikingly predictable:
1. Productivity-boosting infrastructure emerges.
2. That infrastructure becomes widespread, forming platforms and distribution layers.
3. Entirely new consumer behaviors explode on top.
We saw this with electricity. We saw it again with internet and mobile.
My view: we've just entered Phase 1 of the AI revolution. And what follows isn't the end of SaaS—it's a much larger market where upgraded SaaS and entirely new AI categories coexist.
Next comes Physical AI (Phase 2).
And ultimately, Consumer AI (Phase 3).
Part 1: The Repeating Pattern of Technological Revolutions
As mentioned earlier, technological revolutions tend to follow a surprisingly consistent sequence: infrastructure boosts productivity → platforms and distribution take shape → entirely new consumer behaviors emerge. This pattern has repeated itself across every major industrial wave.
Electricity & Mass Production
The late 19th and early 20th centuries were defined by one thing: productivity. Electricity, oil, the internal combustion engine, and the assembly line completely redefined how humanity produced goods. Electricity enabled factories to run around the clock, and assembly lines drove production costs down dramatically. The symbol of this era was Ford's Model T, which turned the automobile from a luxury into something households could realistically own.
But the real transformation came next. The problem was no longer production—it was distribution. Railroads, road networks, telephones, radio, advertising, and large-scale retail networks emerged, allowing goods to spread far beyond local markets. For the first time, a truly national consumer market was born.
And then came the turning point: the birth of the brand. Companies like Coca-Cola, P&G, and Unilever transformed technology from a production tool into a force that shaped lifestyles. People no longer just bought products—they bought into ways of living. Mass production led to mass distribution, and mass distribution ultimately created the modern consumer economy.
In the end, the biggest winners weren't the companies that built the production technology. They were the companies that built new consumer habits on top of it.
The Internet & Mobile Era
The internet followed the exact same script. In the early 1990s, the internet wasn't about social media or online shopping. It was about email, databases, ERP systems, and corporate websites—once again, a pure productivity tool for businesses. Its first chapter was about workplace efficiency and information management.
Then the early 2000s arrived. Broadband spread, cloud computing emerged, and smartphones entered the picture. The internet transitioned into its platform phase.
And then the real stars appeared. Amazon reinvented shopping. Facebook reshaped human relationships. Netflix changed entertainment. Uber transformed mobility. Looking back, the biggest companies of the internet era weren't infrastructure companies. Once again, they were companies that changed consumer behavior. The internet was a technological revolution—but its ultimate outcome was the creation of entirely new daily habits.
Part 2: So Where Is AI in This Cycle?
When you look at today's AI landscape, the pattern feels oddly familiar. What are the most common AI use cases right now?
Coding copilots. Marketing automation. Design assistants. Sales support. Workflow automation. Enterprise SaaS. Internal knowledge tools.
None of this feels new. In fact, it looks remarkably similar to email, ERP, and enterprise software in 1998. There's a reason VC capital is flooding into B2B AI: not because this is the destination, but because it's the beginning of the cycle. In other words, AI today is clearly in Phase 1: the productivity infrastructure stage.
Phase 2 of the AI Revolution: Infrastructure Goes Physical
The next question naturally follows: what needs to happen for AI to move into Phase 2?
Think back to the internet. To reach its platform phase, it needed devices—personal computers and smartphones. Once access became ubiquitous, the internet slipped into daily life and platforms emerged on top of it. AI will likely follow the same path. For AI to become a true platform, it needs to be always accessible in everyday life.
This is where Physical AI comes in. Many people's first thoughts are humanoid robots. That may come eventually—but I believe the shift has already begun long before that. AI glasses, wearables, in-car AI, smartphone assistants. We are already living alongside AI through a growing ecosystem of devices. What's fascinating is that this stage may be much shorter than previous revolutions. It took decades for PCs and smartphones to spread. AI, however, is piggybacking on infrastructure we already have: smartphones and cloud. We may be racing through the "AI infrastructure adoption" phase faster than we realize.
Phase 3: The Explosion of Consumer Behavior
So what happens when we reach Phase 3?
History gives a clear answer: once infrastructure is everywhere, consumer behavior explodes. When cars entered households, means of travel changed. When smartphones became ubiquitous, communication, shopping, mobility, and human relationships changed. AI will be no exception. Once AI becomes everyday infrastructure, entirely new consumer habits will emerge. And on top of those habits, entirely new Consumer AI products and services will be born. In every technological revolution so far, the ultimate winners were the companies that reshaped consumer behavior. There's little reason to believe this time will be any different. We may already be standing right at the starting line.
Finishing Thoughts
Lately, all the headlines about Physical AI have led me to one simple question: what comes next? This post started from that thought.
One piece of advice the VC gurus love is: when in doubt, look at history. And after revisiting the history of technological revolutions, I'm increasingly convinced that this time won't be very different.
Consumer AI isn't some distant future. Just as we look at today's Physical AI news and think, "Damn, robots already?", we may soon find ourselves reacting to new products with, "Wait… how is this possible?"—and before we know it, we'll have built entirely new habits around them.
As I mentioned at the beginning, I don't believe AI will replace SaaS. More likely, SaaS and AI will evolve together and coexist. And somewhere at the end of that coexistence lies the true shape of Consumer AI.
This naturally leads me to the next question: what will Consumer AI actually look like?
This post has already grown longer than expected—so I'll save that for the next one.