AI and Bitcoin valuations show the largest divergence in history, with BTC undervalued by 43% versus AI overvalued by 33%

In April 2026, the global capital markets are experiencing a rare narrative divergence. On one side, the artificial intelligence sector continues to attract massive capital inflows, with hardware and infrastructure companies seeing their valuations soar; on the other side, the crypto market has fallen into silence after a cyclical retreat, with institutional participation still extremely low.

The valuation trajectories of these two asset classes are forming an increasingly widening gap — this is not just a price divergence but also reflects a deep misalignment in global capital allocation logic.

Dan Morehead, founder and CEO of Pantera Capital, provided a precise quantitative assessment. On April 29, at an event in New York, Morehead publicly stated that AI stocks are currently fully priced, while Bitcoin still has about 43% undervaluation relative to its long-term historical trend. He bluntly said this is the biggest market divergence he has seen in his career.

The factual basis of valuation divergence

Morehead’s judgment is based on a set of internal quantitative data. According to Pantera’s analysis system, the valuation index composed of major AI companies is currently about 33% above its four-year logarithmic trend line. This data indicates that the pricing of the AI sector has already front-loaded a significant portion of future growth expectations. Morehead commented that, although artificial intelligence is indeed very important and has enormous growth potential in the long term, at present, “the market has already priced in quite a full outlook for its prospects.”

Meanwhile, Bitcoin’s relative valuation shows a completely opposite picture. According to Pantera’s calculations, BTC’s current price level is about 43% below its historical logarithmic trend line. Morehead used a straightforward statement: “Crypto assets are extremely cheap right now.”

This scale of mismatch — AI overvalued by 33%, Bitcoin undervalued by 43% — together forms an overall pricing gap of approximately 76 percentage points. Morehead characterized this as “the largest divergence ever seen in history.”

Behind this valuation divergence, a more fundamental structural issue emerges: for the world’s largest asset management institutions, crypto assets still almost do not appear on their allocation lists.

Morehead repeatedly emphasized in recent interviews a fact that the market has severely underestimated: “Most institutional investors’ positions in blockchain are still 0.0% — literally zero.” In an institutional system managing over a hundred trillion dollars of global financial assets, crypto assets are almost entirely ignored, which itself constitutes an extremely asymmetric allocation pattern.

It is also worth noting that even institutions that have already allocated show highly concentrated participation. MicroStrategy, for example, bought about 56,235 BTC in April alone, while global Bitcoin ETFs increased their holdings by a total of 34,552 BTC during the same period. These figures far exceed the approximately 11,829 BTC of new miner production during the same period — a significant demand-supply gap.

Based on this, Morehead makes a provocative judgment: “This could be the first time in history that ‘smart money’ is the last to enter.” In previous asset booms, institutional investors with informational and capital advantages usually led the way; but in this blockchain-driven asset paradigm shift, institutions are generally absent.

Is AI really too expensive?

If we view the market as a pricing system, the current state of the AI sector warrants a more cautious perspective. Here, it is necessary to clearly distinguish three levels: facts (what has happened), observations (market phenomena), and inferences (possible scenarios). The following will strictly adhere to this classification.

Capital investment in AI infrastructure is accelerating at an unprecedented scale. According to public disclosures, in 2026, the four major global cloud computing and tech giants — Microsoft, Google, Amazon, and Meta — plan a combined AI capital expenditure of about $660 billion, with Google alone planning to invest $185 billion. Meanwhile, Nvidia reported revenue of $68.1 billion in its fiscal fourth quarter ending January 2026.

There is a noteworthy dislocation between these huge investments and market returns and expectations. Since 2026, the so-called “Magnificent Seven” tech giants have collectively underperformed the S&P 500 index. On April 28, a report about OpenAI’s internal revenue and user targets missing expectations triggered a chain decline in the AI infrastructure sector — Oracle fell about 3%, while Nvidia, Broadcom, AMD, and other chip stocks also declined.

The AI industry is transitioning from the “building phase” to the “adoption phase.” During the building phase, data centers, chips, and cloud infrastructure providers are the biggest beneficiaries; but as it moves into the adoption phase, the market will begin to verify the actual productivity improvements at the end-user level. If the pace of AI commercialization is slower than the implied growth embedded in existing capital expenditure commitments, current high valuations could face correction pressures.

Returning to Morehead’s valuation comparison, the AI index is 33% above its four-year logarithmic trend, which aligns closely with the industry background described above. This does not mean there is a bubble in AI — quite the opposite, as there is solid demand underpinning it — but it does indicate that the current prices of AI stocks already incorporate a considerable degree of optimism.

This reasoning is consistent with Morehead’s statement: he believes AI is very important and has enormous long-term growth potential, but in the short term, “market prices have already fully reflected these positives.” Conversely, crypto assets are in a very different valuation state — pessimistic expectations, lack of capital inflows, and low attention.

Structural causes of institutional absence

The near-zero allocation of institutions to crypto assets is not caused by a single factor. It is a complex picture woven from macro liquidity shifts, geopolitical and policy disruptions, relative valuation attractiveness, and the structural phase of the four-year halving cycle.

On the liquidity front, the relationship between Bitcoin and global M2 is being re-evaluated. As of February 2026, US M2 money supply reached $22.667 trillion and continued to expand. However, Bitcoin’s price has not risen in tandem as traditional liquidity models would predict, showing a significant divergence. Research indicates that M2, as a monthly stock indicator, often takes several months for its transmission effects — via credit expansion and capital flows — to influence crypto assets. Meanwhile, a strengthening dollar and geopolitical risks, which are fast-moving variables, can suppress liquidity benefits in the short term.

Geopolitical disruptions are also significant. In April 2026, tensions in the Middle East pushed Brent crude oil prices up to $115 per barrel, directly raising inflation expectations. The Federal Reserve’s rate cut expectations were reduced from 50 basis points to 25 basis points, and the dollar index strengthened accordingly. For institutional investors, a stronger dollar means tighter global risk appetite, with funds preferring to flow back into dollar assets rather than reallocating into crypto. Under this scenario, even with ongoing M2 expansion, short-term risk asset performance remains under pressure, and institutional willingness to allocate to crypto diminishes.

From a relative valuation and capital expenditure comparison, the four major tech giants’ combined annual capital expenditure of nearly $700 billion creates a significant capital absorption effect — making AI a more attractive and familiar growth story for institutions. In contrast, crypto assets, lacking a widely accepted valuation methodology, are naturally marginalized in institutional asset allocation decisions.

Meanwhile, Bitcoin’s own four-year halving cycle also plays a structural role. Morehead states plainly: “The four-year cycle is real.” Based on the patterns from the previous three complete cycles, Bitcoin reached a cyclical high in the second half of 2025 and is now in a retracement and bottoming phase. The current pullback from the high is about 50%, which is milder than the 85% retracements seen in previous cycles. Still, the bottoming process may take six to eight months. For institutions seeking short-term performance, this presents a substantial participation barrier.

However, from a longer-term perspective, Morehead offers a more structural view: Bitcoin has reached what he calls “escape velocity” — its status as a globally scarce asset is now sufficiently solid that “no factor can derail this process.” Under this framework, the current zero position of institutions is not a negation of Bitcoin’s value but rather a reflection of a historic lag in allocation.

Multi-scenario evolution

Based on all the facts, data, and logical analysis above, here are three possible market evolution paths. It is important to clarify: these are speculative scenarios, intended to provide a systematic thinking framework, not definitive predictions.

Scenario 1: Mean reversion of valuations

Under the assumptions of marginal liquidity improvement and easing geopolitical risks, institutional capital begins to systematically reallocate into crypto assets. Bitcoin ETF fund flows accelerate again, and institutional holdings start to gradually increase from near-zero levels. During this process, Bitcoin may experience a valuation recovery relative to AI stocks, which had previously been excessively priced.

Logic: Market narratives tend to revert after extremes. When the valuation gap between two sectors reaches the largest in history, the incentive for capital to rotate from overvalued assets to undervalued ones becomes strong. Historical experience shows that after excessive concentration in a sector, capital tends to seek rebalancing. Morehead’s characterization — “the largest valuation divergence in history” — itself suggests a higher probability of mean reversion statistically.

Scenario 2: Continued divergence intensifies

Another possibility is that the current divergence persists or even widens. If AI continues to deliver unexpected breakthroughs in productivity and enterprise applications, its high valuation could be fundamentally supported. Meanwhile, the crypto market, constrained by the four-year cycle and lacking new capital inflows, remains in a low-volatility range.

Logic: The scale of capital investment in AI is enormous, and technological iteration is rapid. Even if some indicators underperform, the overall demand for computing power and model capabilities continues to grow. If the transition from construction to adoption phase proceeds smoothly, valuation support will strengthen. Meanwhile, the four-year cycle in crypto has historically held — Morehead himself admits that bottoming may still take 6 to 8 months.

Scenario 3: Systemic re-pricing triggered by macro shocks

The third scenario involves external shocks. If global macroeconomics face unexpected negative shocks — such as a resurgence of inflation, Fed reversing easing policies, or geopolitical conflicts escalating — both AI stocks and crypto could undergo a systemic risk re-pricing.

Logic: Although Bitcoin’s correlation with global M2 has decreased in the short term, long-term liquidity remains a key structural driver. Data shows Bitcoin’s cumulative gains since 2013 are about 700 times, while the liquidity of the world’s five major currencies has grown roughly 100 times. Under extreme stress, crypto’s volatility could be amplified due to its weak institutional base, and high valuations in AI could be revised downward due to profit expectations.

Conclusion

The market divergence precisely indicates that capital is making polarized choices between two clear narrative tracks. AI embodies the most optimistic outlook on technological progress, while Bitcoin reflects deep concerns about the fragility of the monetary system. These narratives are not mutually exclusive — as Morehead states, “There is no world where AI is important and crypto is not involved.”

He further points out that, in the long run, AI and blockchain will converge, and Pantera itself continues to increase investments at this intersection.

When one narrative’s optimism is fully priced in and the other’s pessimism is severely undervalued, the valuation divergence itself becomes a market signal worth cautious attention. This is not a prediction of future asset directions but a sober measure of current global capital misallocation.

According to Gate’s market data, as of April 30, 2026, Bitcoin’s price is $75,693.50, down 2.01% in 24 hours, with a market cap of approximately $1.49 trillion and a market share of 56.37%. Short-term volatility persists, but the structural discussion around the long-term value of crypto assets may have just entered the core agenda of global institutional investors.