The AI Trade Is Pulling Capital Out of Crypto — and the Two Markets Are Now Moving Together
Retail traders and hedge funds have been selling bitcoin and other digital tokens to move their money into artificial-intelligence stocks, the Wall Street Journal reported on August 10, 2026. The article, by WSJ reporter Vicky Ge Huang, documents a shift in capital that has been building through 2026 and now has a name in the market: the AI trade is stealing crypto's thunder.
Bernstein flagged the outflow pattern in June 2026, reporting that bitcoin ETF inflows had slowed sharply as retail investors flocked to AI-related assets. An ETF, or exchange-traded fund, is a basket of assets that trades on a stock exchange like a single share; bitcoin ETFs had become the main way for both institutions and individuals to gain exposure to bitcoin without holding it directly. The ETF channel saw demand cool even as spot bitcoin — the actual token bought and sold on crypto exchanges — remained available. The Bernstein findings, reported via CoinDesk, established the pullback well before the WSJ's August reporting tied the trend to an active rotation rather than simple risk-off behavior — that is, investors pulling back from all risky assets at once.
Yahoo Finance added another layer in June, reporting that insiders were selling a chip stock that had risen more than 200% in 2026 while making rare buys at a crypto-linked name. Insiders are corporate executives or large shareholders who buy or sell their own company's stock; their trades are publicly disclosed. The insider activity suggests that at least some sophisticated market participants viewed the gap between AI stock prices and crypto prices as stretched, even as the broader retail crowd was moving in the opposite direction.
Then the correlation tightened. As August 2026 began, Bitcoin and major cryptocurrencies faced renewed downside pressure following a cooling in AI stocks, Crowdfund Insider reported on August 2. The pullback in AI equities, which had been drawing capital away from crypto, transmitted downward pressure back into digital assets. Capital that had left crypto for AI exposure was not sitting in cash; it was concentrated in a single thematic basket, and when that basket wobbled, the effect rippled through the assets it had drained.
Separately, CNBC reported on July 28, 2026, that brokerages, startups, and retail investors are building AI agents designed to oversee portfolios and automate investing tasks around the clock. The infrastructure layer for AI-driven trading is being built out in parallel with the thematic rotation, which could deepen the integration between AI and crypto markets rather than simply shifting capital from one to the other.
An academic study using eToro trading data, published in the Journal of Financial Economics, found that retail traders are contrarian in stocks and gold — meaning they tend to buy when prices fall and sell when prices rise — but follow a momentum-like strategy in cryptocurrencies, buying when prices are rising and selling when they fall. That behavioral asymmetry matters here: if retail participants were chasing momentum in crypto and have now found a faster-moving momentum vehicle in AI equities, the rotation is consistent with their documented preference pattern, not a departure from it.
The broader context here is that this is not a clean rotation between uncorrelated asset classes. The WSJ's framing captures the direction of capital flow, but the August pullback in crypto following an AI-stock cooling suggests the two trades have become positively correlated at the retail-flow level. When the marginal dollar is allocated by the same participants, using the same momentum-driven playbook, the destination asset inherits fragility from the source.
For portfolio managers, the practical implication is that crypto's beta to AI sentiment has risen through 2026. Beta measures how much an asset's price moves in response to another asset or market; a higher beta means crypto now swings more in tandem with AI stock sentiment. Models that treated digital assets as a diversifier against tech equity risk may be stale. The Bernstein data on ETF inflow deceleration provides a quantitative anchor for the flow reversal, but the Crowdfund Insider report on the August contagion effect is the more operationally relevant signal: the correlation is live, and it transmits in both directions.
The insider activity Yahoo Finance flagged, with sales of a surging chip stock paired with rare buys at a crypto-linked name, is a contrarian marker worth watching. Whether that presages a mean reversion in the AI-to-crypto spread — a return of prices toward their historical average — or simply reflects idiosyncratic positioning is not yet determinable from the available data. What is clear is that the capital-cycle dynamics between these two thematic trades have tightened materially, and the infrastructure being built around AI trading agents, per CNBC's reporting, will likely embed that correlation further rather than loosen it.
