When a founder who has helped build a $134 billion data and AI powerhouse warns that the market is losing its grip on reality, it is worth listening. The spectacle of startups raising billions of dollars with effectively $0 in revenue has become a defining feature of the current artificial intelligence boom, and it is starting to look less like healthy risk-taking and more like a classic bubble pattern.
At the center of this tension is Databricks CEO Ali Ghodsi, whose company has ridden the AI wave while he publicly questions whether the sector’s funding frenzy is sustainable. His critique captures a broader unease among business and technology leaders who see both transformational potential and unmistakable signs of excess in the way capital is chasing AI.
The $134 billion insider calling the AI boom “insane”
Ali Ghodsi is not a skeptic shouting from the sidelines, he is the Databricks CEO who has watched his own company climb to a valuation of $134 billion while warning that parts of the market have become “Insane.” In recent remarks, he described an AI “Bubble” dynamic in which investors are pouring money into companies that have little more than a slide deck and a promise, and he has gone so far as to suggest a 12 “Month Warning” for the most overheated corners of the sector. That kind of language is striking coming from a leader whose business sits at the center of the AI infrastructure stack and has directly benefited from the surge in demand for data and model tooling, yet Ghodsi has argued that the funding environment has “broken” traditional discipline even as he participates in it, a tension that underscores how distorted incentives have become for founders and investors alike, as detailed in one analysis of Ali Ghodsi, CEO of $134 billion Databricks.
That insider status matters because Databricks is not a speculative bet on a single model or app, it sells the plumbing that underpins data and AI projects across industries, from financial services to retail. When someone in that position says the market is “Insane,” he is not just talking about meme stocks or fringe tokens, he is pointing to late-stage rounds where valuations are set on the assumption that any company with “AI” in its pitch will eventually justify a multibillion-dollar price tag. The fact that he is issuing a 12 Month Warning while still raising capital himself highlights a paradox that defined past bubbles as well, where the most sophisticated players recognize the excess but feel compelled to keep playing until the music stops.
Billions with $0 revenue: how the bubble logic works
The most jarring expression of this moment is the willingness of investors to back AI startups with billion-dollar checks even when those companies have essentially $0 in revenue and unproven business models. In private conversations, founders describe term sheets that arrive before a product is live, with valuations justified by vague references to “total addressable market” rather than any concrete path to cash flow. That pattern echoes the late stages of the dot-com era, when companies with no profits and barely any users were treated as inevitable winners simply because they were early to a hot category, and it is precisely the kind of behavior Ghodsi has criticized when he talks about capital being “poured” into untested ideas.
At events like Fortune Brainstorm AI in San Francisco, where Speaking slots are dominated by executives racing to showcase their latest generative features, Ghodsi has blasted the trend of investors chasing hype instead of fundamentals. He has argued that the rush to fund every new model or agent platform, regardless of traction, is distorting how founders think about building durable businesses, a concern echoed in coverage of how Speaking at Fortune Brainstorm AI in San Francisco, Ghodsi framed the current environment. When capital is this abundant and this forgiving, it rewards storytelling over execution, and it encourages companies to prioritize headline-grabbing model releases over the unglamorous work of building reliable products that customers will actually pay for.
Wall Street’s split verdict: bubble, “air pocket,” or evolution?
While Ghodsi is comfortable using the word “Bubble,” other high-profile voices are more cautious, even as they acknowledge that valuations have stretched. In one widely discussed debate, The AI leaders including OpenAI CEO Sam Altman, Bill Gates, and Peter Thiel weighed in on whether the current cycle is a true bubble or something more nuanced, with some arguing that what markets are experiencing is closer to an “air pocket” in which prices may correct without derailing the long-term trajectory of the technology. That framing reflects a belief that, although certain companies are clearly overvalued, the underlying advances in large language models, chips, and data infrastructure are substantial enough to justify a significant re-rating of the entire sector, a view captured in reporting on how The AI bubble debate: 17 business leaders, from Sam Altman to Bill Gates to Peter Thiel, weigh in.
I see this split verdict as a sign that the market is trying to reconcile two truths at once: that AI is a genuine platform shift and that investors are still capable of overpaying for that shift in spectacular fashion. On one side are those who look at the sheer scale of investment and the proliferation of copycat startups and conclude that a painful correction is inevitable. On the other are optimists who argue that, even if some valuations fall, the aggregate value created by AI over the next decade will dwarf the losses. The tension between “Bubble” and “air pocket” language is less about data and more about risk tolerance, and it leaves founders and employees navigating a landscape where both massive upside and brutal whiplash are plausible outcomes.
Databricks as a case study in froth and fundamentals
Databricks itself illustrates how hard it is to separate genuine business momentum from market froth. Earlier this year, the company moved to raise a $4B Series L at a valuation of $134, up from $100 only a few months prior, a leap that underscores just how aggressively investors are bidding up perceived category leaders. That kind of step-change in private valuation, captured in coverage of the Databricks Series L, reflects both strong demand for its data and AI platform and a willingness among late-stage funds to pay almost any price for exposure to the infrastructure layer of the AI stack.
From my vantage point, Databricks sits at the intersection of two powerful forces: the real need enterprises have to wrangle their data and build AI applications, and the speculative impulse that treats any company enabling that shift as a must-own asset. The company’s ability to command a $4B Series L at a $134 valuation, after previously being valued at $100, suggests that investors see it as one of a small handful of platforms that will define how AI is built and deployed. Yet the speed of that repricing also raises the question Ghodsi himself is asking about the broader market, namely whether even strong businesses are being valued on assumptions that no operating performance can realistically meet once the cycle turns.
Is this really a bubble, or just the next tech supercycle?
Not everyone accepts the “Bubble” label, and some data-driven commentators argue that what looks like mania is actually a familiar pattern in major technology shifts. One widely shared analysis framed the current AI surge as part of a long arc in which Everyones calling AI a bubble, but the underlying metrics look more like the early innings of cloud or mobile than the late stages of a speculative blowoff. The author contrasted current AI valuations and revenue multiples with those seen During the dot-com peak, arguing that, while pockets of excess are obvious, the overall market is not as stretched as the loudest critics suggest, a perspective laid out in a post that begins, “Oct, Everyone’s calling AI a bubble. The data tells a completely different story. 𝗧𝗛𝗘 𝗖𝗢𝗠𝗣𝗔𝗥𝗜𝗦𝗢𝗡: During the,” which is accessible through Everyone’s calling AI a bubble.
I find that argument compelling up to a point, especially when it emphasizes that every major platform shift, from the internet to smartphones, has involved a phase where capital floods in faster than revenue can catch up. The key distinction, however, lies in how concentrated today’s bets are in a relatively small number of foundational model companies and infrastructure providers, and how many smaller players are being funded on the assumption that they will be acquired rather than ever becoming profitable on their own. That structure can amplify volatility, because if the giants slow their spending or consolidation takes longer than expected, the startups with $0 revenue and sky-high burn rates will have little cushion. Whether we call that a Bubble or a supercycle, the practical advice for founders and investors is the same: treat the current abundance of capital as temporary, build products that customers will pay for even in a downturn, and remember that the loudest hype cycles often end just as the most durable businesses are quietly getting to work.
More From TheDailyOverview
Grant Mercer covers market dynamics, business trends, and the economic forces driving growth across industries. His analysis connects macro movements with real-world implications for investors, entrepreneurs, and professionals. Through his work at The Daily Overview, Grant helps readers understand how markets function and where opportunities may emerge.
