AI Bubble Vs Dot-Com Bubble: Are Today’s Valuations Repeating History or Writing a New Story?

Synopsis: The AI boom mirrors aspects of the dot-com bubble, with soaring valuations and concentrated gains. While major firms show strong fundamentals, many startups remain speculative. The key question persists: Is this history repeating or a new AI-driven story?

The swift rise of generative artificial intelligence and the sharp increase in technology investments have reignited a familiar debate: is the current AI boom beginning to resemble the dot-com bubble of 1999-2000? Advocates of the bubble argument highlight surging equity valuations, pointing to examples such as NVIDIA’s stock climbing roughly 1,300 percent since late 2022, alongside lofty private market valuations, including OpenAI at over USD 300 billion and Databricks at USD 62 billion.

They also cite a wave of heavily marketed AI products that often fade as attention shifts to the next major announcement. On the other hand, skeptics argue that artificial intelligence represents a durable, foundational technology rather than a short-lived trend.

They note its widespread adoption across enterprises, with dozens of AI implementations occurring daily, as well as strong near-term revenue visibility across several parts of the ecosystem. In reality, the picture is more nuanced: while certain segments and stocks appear driven by speculative excess, others are supported by tangible business models and measurable economic value.

Understanding The Dot-Com Bubble

To evaluate the similarities, it is useful to briefly revisit the dot-com period. Beginning in the mid-1990s, the commercialization of the internet triggered a powerful wave of speculation. Internet startups multiplied rapidly, many built around the promise of fast user growth rather than near-term profitability. Capital from venture investors and public markets flowed freely into anything carrying a “.com” label, while conventional valuation yardsticks such as earnings, cash flow, and balance-sheet strength were often brushed aside. 

A familiar pattern emerged: internet companies with little or no revenue went public at lofty valuations, raised substantial capital, and spent aggressively on customer acquisition and expansion. Stock-option-rich employees turned into overnight paper millionaires, further fueling the speculative momentum.

The boom reached its apex in March 2000. The Nasdaq Composite Index, dominated by technology stocks, climbed to around 5,048, nearly triple its level in early 1998. Valuations became stretched across the sector, with the Nasdaq-100’s forward price-to-earnings ratio approaching 60 times at the peak. 

Large technology leaders such as Cisco, Intel, Yahoo, and Sun Microsystems traded at exceptionally high price-to-sales multiples. Importantly, profitability was rare across the dot-com universe; estimates suggest that only about 14 percent of internet companies were generating profits at the time. Reflecting on the episode years later, investor Rob Arnott captured the prevailing mindset, observing that “the narrative was correct, but the market bet that narrative would play out a lot faster than it ultimately did.”

The turning point came in early 2000, when the US Federal Reserve began raising interest rates. Market sentiment quickly reversed, and between March 2000 and October 2002, the Nasdaq collapsed from 5,048 to 1,139, wiping out nearly all of its gains. Thousands of dot-com startups shut down or were acquired at deeply distressed valuations. 

The crash inflicted significant losses on investors and served as a reminder that traditional fundamentals, assets, profits, and cash flows, could not be ignored indefinitely. Retrospective assessments of the period consistently highlight extreme optimism, concentrated ownership, and reckless valuation assumptions as defining features of the bubble. Venture capital investment in the US reached a peak of USD 112.3 billion in 2000, while median price-to-sales ratios among technology companies surged to extraordinary levels, with Cisco at one point trading near 200 times sales.

Taken together, these data points underscore how the late-1990s environment was marked by overheated technology valuations and widespread irrational exuberance, ultimately ending in a sharp and painful correction. The key question today is whether the AI-led markets of 2022-25 exhibit the same characteristics.

Understanding The AI Boom

Since late 2022, the rise of generative artificial intelligence, driven by platforms such as ChatGPT, Anthropic’s Claude, and Google Gemini, has triggered intense interest across sectors. Companies and investors alike have been scrambling to position themselves early and capitalize on the accelerating AI wave.

AI Spending Surge 

Multiple research firms project a dramatic expansion of the AI market. IDC, for instance, estimates that global annual expenditures on AI, including hardware, software, and services, will more than double, reaching approximately USD 632 billion by 2028. Worldwide, data centers are ramping up deployments of AI accelerators to meet growing demand. According to Bain & Co., supporting the generative AI revolution will require companies to generate combined revenues of roughly USD 2 trillion by 2030 to cover computing costs, yet they are projected to fall short by about USD 800 billion. Bain describes this shortfall as a significant “AI compute funding gap.”

Venture capital trends mirror this rapid growth. AI startups are attracting record levels of investment. In the first quarter of 2025 alone, AI ventures raised USD 73.1 billion, accounting for nearly 58 percent of total global VC funding. Over the first half of 2025, PitchBook reports that AI startups secured USD 104 billion of the USD 205 billion total VC investment worldwide, representing 53 percent of all funding. In the United States, this concentration rises to roughly 64 percent of venture capital. 

By comparison, no single sector commanded such dominance in VC funding during the 1999-2000 dot-com era. This extraordinary concentration reflects both excitement about AI’s transformative potential and the fear of missing out. Databricks CEO Ali Ghodsi highlighted this fervor in late 2024, noting that AI fundraising seemed to have reached “peak” levels, his company raised USD 10 billion at a USD 62 billion valuation, more than twice its original target.

Rapid Adoption of AI

By the end of 2024, surveys indicate that 78 percent of companies were actively using AI. Data from McKinsey and Stanford show adoption rising sharply from around 50 percent in 2022 to over 70 percent across various business functions. Major corporations, including Walmart and Goldman Sachs, are deploying AI solutions across multiple units, reflecting broad and genuine engagement with AI technologies well beyond pilot or hype projects.

On the consumer side, applications like ChatGPT reached 100 million monthly users within just two months of launch, an unprecedented pace for any consumer app. Additionally, billions of people now regularly interact with AI through voice assistants, recommendation systems, or image-generation tools. By comparison, in 2000, the internet, including web and email, had neither such rapid global uptake nor such widely adopted consumer “killer apps,” with just over half of US households online and social media and mobile technology still in their early stages.

Revenue & Profitability Gap

Companies at the core of the AI ecosystem are reporting substantial sales expansion. NVIDIA, for instance, is projecting nearly USD 120 billion in revenue for fiscal 2025, almost double its prior-year figure. Meta Platforms (formerly Facebook) has increased profits by leveraging AI to enhance ad targeting, while major cloud providers such as AWS, Azure, and Google Cloud report robust uptake of AI services. By contrast, most pure-play dot-com companies around 2000 generated minimal revenue and recorded significant losses.

Large AI-focused firms display markedly strong margins. NVIDIA’s net margin of 53.4 percent, for example, surpasses those of even the FAANG giants, highlighting the unique economics of AI hardware. Yet, many AI-specialist startups continue to burn cash. Firms like OpenAI (were it publicly listed) post relatively modest revenues compared with their high valuations. This situation echoes the dot-com era pattern, where a handful of major players thrived while numerous startups operated at losses.

The Verdict: AI Bubble Vs Dot-Com Bubble

Valuations 

During the dot-com era, technology stocks traded at extremely high multiples, Nasdaq-100 forward P/E reached around 60x, compared with roughly 26x for leading AI-related stocks today. Even major hardware firms such as Cisco saw price-to-sales ratios near 200x. In contrast, current AI-focused companies generally exhibit more moderate valuations. For example, NVIDIA’s price-to-sales ratio peaked at about 50x in 2024, which, while elevated, is not unprecedented relative to its growth trajectory. This suggests that today’s stock prices reflect at least some underlying fundamentals.

Business Models

During the dot-com era, the market was dominated by young startups, many with no profits and untested business models, over 85 percent of which failed once investor enthusiasm waned. Today, however, the leading players in AI are largely established firms with decades of operating history, such as Microsoft, Google, Amazon, and NVIDIA. These incumbents already generate substantial, sustainable revenues and often maintain strong cash flows, providing a buffer if AI investments underperform.

The rapid rise in AI startup valuations evokes memories of the dot-com period. A late-2025 Reuters report highlighted that investors continue to pour capital into AI startups despite modest revenues. Per-employee valuations in the AI sector have reached unprecedented levels, with no clear parallel from the dot-com era, given that those companies typically employed far fewer staff relative to their valuations. 

Many AI startups still lack well-established revenue models, such as generative content companies relying on enterprise licenses or ad-supported consumer apps with uncertain scalability. Consequently, similar to the dot-com period, the greatest risk may reside with these newer entrants rather than the large, established incumbents.

Unlike the 1990s, today’s AI investment surge is heavily capital-intensive. Significant funds are being allocated to physical infrastructure, including data centers, GPU clusters, and fiber networks. This has dual implications.

On the positive side, heavy investment in tangible assets suggests that durable infrastructure is being built rather than merely fueling marketing hype. On the other hand, if demand projections fail to materialize, these assets could remain underutilized, negatively affecting returns. The Business Times notes that financing structures for AI build-outs, including complex debt arrangements, are raising eyebrows, indicating financial strain beyond pure equity investment.

Market reaction

Both the dot-com era and today’s AI-driven market exhibit high concentration in a few dominant companies. In 1999, four firms, Microsoft, Cisco, Intel, and Dell, accounted for a substantial portion of the Nasdaq. Similarly, by 2025, just three AI and tech giants are driving most of the S&P 500’s gains. NVIDIA alone contributed roughly 32 percent of the S&P’s gains through mid-2024. This concentration means that the performance of the broader market is heavily dependent on the fortunes of a small number of key players, and a stumble by any of these firms could significantly impact overall market performance.

The dot-com period experienced hundreds of IPOs and extreme year-to-year volatility. Between 2023 and 2025, there has been a surge in AI-related IPOs, including companies like C3.ai and SoundHound, along with notable AI-driven stock rallies. However, regulatory measures and market conditions are somewhat tighter, and we have not yet witnessed the massive overnight wealth losses seen in 2000.

Significant swings have occurred, such as NVIDIA’s stock rising tenfold before retracting, or crashes in biotech- and AI-linked stocks. Much of the public-market risk is currently deferred because many AI unicorns remain privately held; their eventual IPOs could generate sudden market shocks.

In the dot-com crash, investor losses in equities triggered a broad bear market. Today, central banks are more alert to deflation risks, and technology represents a larger share of household and institutional wealth. While an AI-sector correction might initially be concentrated in tech stocks, the interconnected nature of modern financial markets means that fallout could potentially extend more broadly.

Expert Sentiment 

A Bank of America survey conducted in October 2025 found that 54 percent of fund managers considered AI-related equities to be “in a bubble” or overvalued, while 60 percent labeled global equities more broadly as overpriced. A Quartz summary reported similar findings, noting that 54 percent of respondents believed tech stocks, driven largely by AI hype, were priced too high. BofA analysts further highlighted AI as the top “tail risk” for investors, outweighing concerns about inflation and geopolitical tensions.

Several leading figures in the AI space have publicly cautioned against excessive optimism. OpenAI CEO Sam Altman, following a major fundraising round, warned that investor enthusiasm may be outpacing the technology’s proven capabilities. Alibaba’s CEO halted an AI hiring surge to prevent overcapacity.

Notably, Databricks CEO Ali Ghodsi, while raising a record USD 10 billion round, explicitly referred to it as “peak AI bubble.” These insider warnings contrast with earlier phases of AI hype, such as sensational claims about imminent artificial general intelligence, suggesting a more cautious sentiment at the apex of the cycle.

Industry economist Torsten Sløk of Apollo Global argued in 2023 that the top ten companies in the S&P 500 were more overvalued than the top ten firms were during the mid-1990s tech bubble. Investing pioneer Rob Arnott told the Financial Times that AI hype represents “a classic example of a big market delusion… just like the dot-com era.”

Meanwhile, other experts, including those from McKinsey and Bain, warn that the AI sector could still face a “trough of disillusionment” or a period of consolidation within the hype cycle.

Opinions in the venture capital community are divided. Surveys indicate that while some investors believe current AI valuations are unsustainable, others remain optimistic about long-term potential.

A 2024 poll of technology-focused VCs found that roughly 40 percent saw an “AI bubble” forming, whereas 45 percent disagreed, highlighting the lack of consensus. This split reflects both the enthusiasm of investors riding the AI wave and the caution of a sizable faction concerned about overvaluation.

-Manan Gangwar

Disclaimer

The views and investment tips expressed by investment experts/broking houses/rating agencies on tradebrains.in are their own, and not that of the website or its management. Investing in equities poses a risk of financial losses. Investors must therefore exercise due caution while investing or trading in stocks. Trade Brains Technologies Private Limited or the author are not liable for any losses caused as a result of the decision based on this article. Please consult your investment advisor before investing.

The post AI Bubble Vs Dot-Com Bubble: Are Today’s Valuations Repeating History or Writing a New Story? appeared first on Trade Brains.