Are we in an AI bubble? Yes... and no. Both can be true. We can be in a bubble AND massive value can still be created. These aren’t mutually exclusive. AI will have a much greater economic impact than the internet or mobile - and both created tens of trillions of dollars in value. Part of that is because AI is layered on top of those technologies. The benefits compound. The risk isn’t whether we’re in a bubble, but the timing mismatch between when people expect returns and when they actually show up. If you take a ten-year view, there’s no bubble at all. But if you’re focused on the next few quarters, you might see valuations correct before fundamentals catch up. As Amara’s Law reminds us, we tend to overestimate the short term and underestimate the long term. Will OpenAI hit its revenue targets by 2029? Hard to say. But over the next decade, the combined output of OpenAI, Gemini, Microsoft Copilot, Anthropic, and Perplexity will absolutely create transformative value. That will be true even if we’re in a bubble today. The capital intensity of AI will surpass that of the early internet, but that’s not irrational. Disruptive cycles have always required heavy investment. Nvidia and others are spending now to build the foundations of the AI economy. Rather than worrying about whether we're in a bubble, founders should focus on the fundamentals: Do you have unique insight? Are you building something that creates genuine value? Do you understand your customer’s workflow better than anyone else? Startups that get those things right will capture the upside, bubble or not.
Ashu Garg - exactly my sentiment. Having been both on the buy side in large enterprises, and now as a Co Ffunder of Rezolve.ai - I have seen the mismatch in timing between what's possible and how companies actually adopt and use it first hand. The massive transformational impact of AI is a matter of when not if.
This is such a clear take. Every innovation cycle looks like a bubble if you’re only watching quarterly returns. But founders who build on long-term fundamentals aren’t speculating, they’re compounding. The noise is in valuation swings the signal is in adoption curves. The real question isn’t “are we in a bubble,” it’s “who’s building something that outlives it
Every tech whether it's internet, desktop, mobile, cloud, etc., took a decade to materialize and become part of the tech eco system. AI is going to be no different and the industry knows it. The problem is the frothy and insane valuations with little or no revenue. In short, AI is a tech that'll boom (like that every tech that came before it) and the valuation is the bubble which pops (for e.g, Internet/dot-com) same as it did before followed by the amazing and fruitful transformation. The sooner it pops is generally good because that's when focus will be on real world problems that needs solving!
I recently wrote a LinkedIn article about this. The key takeaways: If we are in a bubble, which we probably are, it’s an innovation bubble, not a financial bubble. Financial bubbles are the destructive bubbles, where prices rise untethered from real value, and the burst erases wealth without leaving much behind. Innovation bubbles are different. They arise when capital and talent pour into an emerging field too quickly. The result looks irrational in the short term but proves transformative over time. While a lot of capital will be lost and a lot of companies will go under, the accelerated learning will leave real value and innovations behind.
Ashu Garg , I feel Gen AI value delivery is experiencing a Schrödinger’s cat position. Many of your articles emphasized on a long term strategy and executing plan. In a world of consistent change , weather building a strong AI wrapper or not AI powered solutions , a customer centric approach will pave the way for a successful products.
The real AI revolution is not what we have now. The real AI revolution will be when AI gets efficient enough that everyone can do, on their laptop, what now takes an entire server room to do. In 1960, an entire server room had approximately the same capacity as today's smartphone. The future possibilities are far more than what even innovators of today are working on.
Pranab Ghosh
4w
If you are referring to GenAI, there is no such evidence so far. The only economic impact is negative. There is no sign of significant returns from trillions invested. “AI will have a much greater economic impact than the internet or mobile - and both created tens of trillions of dollars in value.”
Ajay N.
4w
I am right, I am wrong. Context darling.
Vin P.
4w
Yno!
This is going to be exactly same as internet / dot com bubble, bust and then long term value creation. When the internet bubble burst ; shaky business models were impacted. But the infrastructure laid to the Amazons and Googles creating real value Smart enterprises are looking to AI as long term investment; combining intimate knowledge of customers workflows, business requirements and customer data with reliable, explainable AI with no-rush rollout and plenty of guardrails. These will contribute to real value creation