How Long Will the AI Boom Continue? The #1 Question for Crypto Investors

The current discussion revolves around the duration and nature of the AI boom, particularly whether it constitutes an AI bubble and its implications for crypto assets. The correlation between crypto and the NASDAQ has never been higher than it is in 2026, with crypto prices seemingly being pulled upwards by a strong tech stock market, especially the NASDAQ. This raises the critical question of how long the AI boom will last and its impact on crypto investors. The term "bubble" often evokes mixed feelings. Some dismiss it as meaningless, while others, like author Burn Hobart, argue that bubbles can be historically valuable. However, for this discussion, a bubble is defined through the lens of Carlota Perez's framework of technological revolutions. This framework outlines phases: eruption, frenzy, turning point, synergy, and maturity, closely mirroring the Gartner hype cycle with its trigger, inflated expectations, trough of disillusionment, slope of enlightenment, and plateau of productivity. Every major technological revolution, including the internet, radio, electricity, railroads, and automobiles, has followed this pattern, and crypto has experienced multiple such cycles. The challenge lies in identifying the exact position within this cycle, as the duration and intensity of each phase are unpredictable. The AI boom's eruption phase can be traced to the release of ChatGPT in late 2022, with 2024 and 2025 likely representing the frenzy period. The current sentiment suggests we might be in the later stages of this frenzy, though its duration remains uncertain. The scale of AI's impact, whether it's a "six," "seven," "eight," or "nine" on a "technical RTOR scale" (similar to Nate Silver's concept), will determine the slope and longevity of this frenzy. Examining market data provides insight into the current valuation of equities. The Schiller PE CAPE ratio, which is a cyclically adjusted PE ratio averaging the last 10 years and adjusted for inflation, currently stands at 42. This is very close to the peak of 44 seen during the dot-com bubble in 1999 and significantly higher than the 33 in 1929. The only time in history the Schiller PE was higher was in 2000, indicating that stocks are not cheap, and forward returns are likely to be subdued if bought at these levels. However, a contrasting narrative suggests that "this time is different" due to strong revenue growth forecasts. Q1 earnings growth is estimated at 27.7%, significantly higher than the 10-year average of 10.3%. The argument is that unlike the dot-com era, where many companies were unprofitable "fluff," today's AI buildout is financed by highly profitable, well-capitalized hyperscalers. Yet, historical data from 1999 shows that earnings estimates were also ramping up similarly, with Q4 1999 seeing the same expected earnings growth as today, and Q1 2000 even higher at 32.7%. This suggests that strong earnings growth is not unique to the current period and that price often leads fundamentals in hype cycles. Forward PE ratios offer another perspective. The "MAG7" (presumably the Magnificent Seven tech stocks) have a forward PE of 26.7, and the large-cap S&P 500 is around 21. While not at all-time highs (they were higher in 2021), this is because earnings are keeping pace with price increases. This data can support a bullish argument that current valuations are justified by strong earnings. However, the critical question remains: is price leading fundamentals, or are fundamentals leading price? In hype cycles, price tends to lead. S&P 500 forward profit margins are also at historical highs, currently at 15.3%, the highest since 2004. This fuels the narrative of AI-driven productivity and efficiency. However, it's unclear if this margin improvement is solely due to AI implementation, as studies confirming the ROI of AI spend in large enterprises are not yet prevalent. Other factors, such as reduced hiring (as seen in some companies using AI to backfill roles), could also contribute to margin improvements. The concern is whether the current AI spend will yield the expected ROI, and if not, this could shift the narrative and reduce demand. The short-term market movements exhibit frothy conditions. The NASDAQ experienced a 26% rally over five weeks, an event that has only happened eight times since 1971. Historically, such rallies either occur as mean reversions after cycle lows or at market tops. Three of these instances occurred during the dot-com bubble (1998, 1999, and March 2000), suggesting the current rally could be an early, middle, or final blow-off top. This creates a challenging environment for long-term investors, while traders might find opportunities in the volatility. Market concentration levels are also noteworthy. The top 10 AI companies currently account for 40% of the S&P 500, a level comparable to previous major bubbles like the Nifty 50 (1960s) and the Japanese market in the 1980s. While not an exact apples-to-apples comparison, it highlights a significant concentration. Interestingly, not all of the MAG7 are participating equally in the recent rally, with only four returning to all-time highs. This dispersion, where broad market participation is lacking (the equal-weight S&P 500 is not at all-time highs), aligns with patterns observed during past bubbly periods. Examples like Intel (up 200% in 5 weeks) and memory stocks like SanDisk (up 540% year-to-date) and Micron (up 130% since April 1st) illustrate where the current exuberance is concentrated within the AI stack. The mechanics of AI money flow today are driven by enterprise demand. Companies are rushing to integrate AI solutions to improve margins and efficiency, leading to subscriptions and API usage flowing to model and application layers like OpenAI, Anthropic, and Perplexity. These companies, though rapidly scaling revenues (Anthropic's annualized revenue reportedly jumped from $10 million in December 2022 to $45 billion in May 2026), are not always profitable. They, in turn, spend heavily on hyperscalers (Google, Amazon, Meta) for compute, which then flows to chip manufacturers like Nvidia and TSMC. This circular flow is sustained by the continuous improvement of AI models and the FOMO-driven race to implement these solutions. The key risk is a potential slowdown in demand if the ROI is questioned or if cheaper alternatives emerge, leading to an oversupply of compute capacity, similar to the overbuilding of bandwidth during the dot-com era. The dot-com bubble exhibited similar capital flow mechanics: end-user demand for internet access led to telecom companies (AT&T, Verizon) building out fiber optic infrastructure. Too much investment flowed into this infrastructure, overshooting near-term demand. The question for AI is whether a similar overbuild of data centers and compute access will occur. Currently, there's no evidence of a glut of supply, but this doesn't preclude it from happening, especially if price leads fundamentals and market sentiment shifts. Investor positioning appears predominantly bullish, with retail call options exploding to 9 million contracts on a five-day average, significantly higher than the 6 million at the 2021 peak. Hedges have been removed, and the VIX (volatility index) has fallen, indicating increasing market complacency. Credit spreads are tight, suggesting easy access to capital. The end of the dot-com bubble was marked by extreme concentration, which began to break down, and significant dispersion in market performance. A restrictive Fed, hiking rates into an overheating economy, also played a role. While there wasn't a single catalyst, concerns about a recession in Japan and an antitrust lawsuit against Microsoft contributed to a shift in market narrative. The NASDAQ ultimately dropped 78% from March 2000 to October 2002. Factors like future IPOs (similar to token unlocks in crypto) and reduced share buybacks (as MAG7 companies prioritize capex for AI) could impact market structure. The lack of broad market participation, with the equal-weight S&P 500 lagging, also echoes dot-com era patterns. For crypto investors, the current situation presents a dilemma. Bitcoin's correlation with the NASDAQ is at an all-time high in 2026, and Bitcoin typically leads the NASDAQ. If the NASDAQ continues its rally and then crashes, crypto could be pulled along. The best approach involves understanding this correlation. For traditional finance exposure, index funds are a passive strategy. In a bubble scenario, stacking cash and waiting for "fat pitch" opportunities when assets are oversold is advisable. Currently, Bitcoin is at a critical inflection point around its 200-day moving average, which could act as resistance in a bear market. If Bitcoin breaks down while the NASDAQ continues to rally, it could signal crypto markets decoupling and pursuing their own trajectory. Crypto markets are viewed as free markets, less subject to external manipulations seen in traditional finance, offering a different dynamic for investors to navigate.