Three Years Later: How “The Great Reset” became The Great Reinvention
“Let’s go reinvent tomorrow instead of worrying about what happened yesterday.”
- Steve Jobs
Looking back at my January 2023 post about "The Great Reset," I'm struck by both what I got right and what none of us saw coming. Yes, I along with most everyone else predicted the return to fundamentals, the emphasis on unit economics, and the rationalization of valuations. What I didn't fully grasp was that we weren't just resetting - we were standing at the precipice of a complete reinvention of how companies are built, funded, and scaled.
The "bumpy ride" I predicted turned out to be more of a violent shakeout followed by an explosive renaissance. The venture capital industry didn't just correct; it metamorphosed.
The Predictions vs. Reality Check
My 2023 predictions were directionally correct but understated in magnitude:
What I Got Right:
Valuations did rationalize, but only for companies without AI narratives. The bifurcation was more extreme than anticipated.
Due diligence intensified dramatically, but it evolved to focus on AI moats, inference cost structures, and data network effects rather than traditional SaaS metrics.
Governance became critical, particularly around AI safety, data rights, and algorithmic transparency.
Profitability metrics returned with a vengeance—the era of "growth at all costs" definitively ended by Q3 2023.
What I Missed: The speed and scale of AI's impact. While I was writing about rational valuations in January 2023, OpenAI was already preparing GPT-4. The subsequent AI arms race created an entirely new venture paradigm where traditional valuation models became nearly obsolete for AI-native companies.
What’s Ahead For Us In 2026
As we enter 2026, we stand at the edge of a reckoning three years in the making. The venture landscape has been violently reshaped since the 2023 Purge that saw 40% of unicorns accept down rounds, followed by the Great Divergence when AI companies commanded 50-100x multiples while traditional SaaS companies struggled at 5-7x.
Now we've reached the Synthesis Phase - a new equilibrium where every company is judged through the lens of AI leverage. Meaning that investors and markets are no longer asking “Does this company use AI?” but rather “How effectively does this company multiply its capabilities through AI?” It’s the difference between having AI as a feature versus AI as a force multiplier fundamentally embedded in the company’s operations.
But this equilibrium is about to shatter and expose some painful truths. The inference costs that were supposed to plummet never did, leaving thousands of AI companies with broken unit economics. The $800K AI engineer is about to become the $500K web developer of 2001 as AI begins building AI. Of the 15,000+ AI startups founded since 2023, perhaps 500 will survive the year. One major data breach or AI scandal will trigger a regulatory avalanche that will bury the unprepared overnight.
For investors, 2026 won’t be about finding AI opportunities – it will be about avoiding AI landmines. The real returns will come from counter-intuitive bets: micro-unicorns with three-person teams building $100M businesses, second-order effects of AI disruption, and companies solving the problems AI creates rather than competing with it. For founders, the playbook has flipped entirely - your defensible moat can no longer be features that AI can replicate in weeks. It must be based on proprietary data networks, regulatory capture, or deeply embedded human relationships. The winners won't have the best AI (that's already commoditized) but those who understand what AI can't do and where “human-in-the-loop” commands a premium. Building for the post-AI world means focusing on trust infrastructure: hallucination detection, synthetic data validation, and human identity verification in an AI-saturated environment.
The weird truth is that 2026 won't be about building the future - it will be about surviving the present. We're witnessing the “four horsemen of the AI apocalypse” converging: model collapse from AI training on AI-generated content, a privacy reckoning as questionable data practices are exposed, a massive consolidation wave as 90% of AI startups will likely fail, and paradoxically, a flight to human interaction as customers flee over-automated experiences. Fun times for sure.
The innovation industry that emerges from this will be leaner and more disciplined, purged of tourists and AI-washers. The experimental phase is over; the speculation bubble has burst. What remains is the hard work of building real businesses with sustainable economics in an AI-native world where being profitable (or at least not torching cash) isn't just important - it's existential.
The Lesson of the Great Reinvention
Looking back at my 2023 predictions, I was right about the need for a reset but wrong about its nature. We didn't just need better metrics and governance; we needed an entirely new framework for understanding value creation in an AI-native world. The "Amazons, Microsofts, and Googles" that I predicted would emerge from this period? They're here, but they look nothing like their predecessors. They're smaller, more efficient, and more powerful. They're built on AI foundations that didn't exist when I wrote that post and are building for a world where AI is as fundamental as electricity. They're solving problems we couldn't even articulate in 2023: ensuring AI alignment, managing synthetic realities, preserving human agency in an automated world, and creating value in an economy where intelligence is abundant.
As we all look to navigate 2026, the lesson is clear: The pace of change has accelerated beyond traditional planning horizons. The only sustainable strategy is continuous adaptation. Those with conviction and patience will still be rewarded - but now it's conviction in change itself, and patience measured in quarters, not years.
The Great Reset became The Great Reinvention. And we're just getting started.
#ONWARD
