Crypto ETFs lose assets as AI gains

TLDR

Crypto-focused ETFs are seeing significant outflows while AI themed tech and chip funds attract strong inflows, showing a capital rotation rather than a simple risk off move.

1. U.S. spot Bitcoin and Ethereum ETFs have lost roughly $2.32.7 billion over two weeks, with multiple days of net outflows across major issuers.
2. At the same time, AI and semiconductor vehicles like the DRAM memory ETF and AI heavy tech stocks are pulling in billions, with hedge funds at record tech overweights.
3. For crypto holders, this tightens one big institutional demand channel, so ETF flow data, interest rate expectations and any AI plus crypto convergence will be key signals.

Deep Dive

1. Scale Of Crypto ETF Outflows

Recent reports show U.S. spot Bitcoin ETFs saw about $1.26 billion of net redemptions in a single week, their largest weekly outflow of 2026 so far, with Ethereum products also in sustained outflow.CryptoSlates summary puts combined BTC and ETH ETF outflows near $2.7 billion over the last two weeks, while another analysis cites $2.26 billion in Bitcoin ETF redemptions alone.

These outflows come even as total Bitcoin ETF assets remain large, with spot BTC ETFs still holding around the low one hundred billions of dollars and Ether products roughly in the mid teens of billions, according to recent aggregate AUM data. There is also some rotation inside crypto itself: one report notes spot Solana and XRP ETFs drew about $15 million and $22 million in net inflows respectively over the same period.A market recap highlighted this selective altcoin demand.

Analysts link the redemptions to rising bond yields, shifting Federal Reserve expectations and profit taking after prior inflows, all of which reduce appetite for non yielding assets like Bitcoin when yields and alternative trades look more attractive.

2. AI And Chip Vehicles Gaining Assets

While crypto ETFs leak assets, AI themed products are booming. The DRAM memory chip ETF, which gives exposure to AI infrastructure via high bandwidth memory and related semis, has become one of the fastest growing ETFs ever, amassing more than $6.5 billion in assets in its first 27 trading days and surpassing $10 billion within about 30 sessions.A detailed piece notes that puts it among the top U.S. ETFs by year to date inflows.

Hedge fund data cited by Goldman Sachs shows technology allocations at record highs relative to global benchmarks, with the heaviest buying in semiconductor makers and software names tied directly to artificial intelligence.A CoinsKid community summary emphasizes that institutional portfolios are now as overweight tech as at any point in the last decade.

In other words, risk capital is not simply retreating. It is aggressively chasing AI and hardware narratives, and crypto is currently losing that tug of war inside multi asset portfolios.

3. Why It Matters And What To Watch

ETF flows have become one of the clearest gauges of institutional demand for Bitcoin and Ether. Several articles argue that recent outflows are a key factor preventing BTC from holding above prior highs, even though spot prices remain elevated by historical standards.One flow focused update directly links the $2.26 billion two week outflow to Bitcoin stalling below psychological resistance.

Macro is an important driver. Earlier in the year, ETF inflows were supported by expectations of Fed rate cuts. As inflation proved sticky and markets began to price fewer or no cuts, futures pricing shifted toward a higher for longer stance, which favors yield and infrastructure trades over long duration growth assets like crypto and speculative tech.

Conclusion

Capital is not abandoning risk entirely. It is rotating from crypto ETFs toward AI and semiconductor exposure as institutions crowd into what they see as the next structural growth trade. For crypto, sustained ETF outflows can cap upside even in a broadly bullish environment, but the longer term thesis depends more on adoption, regulation and eventual convergence with AI and tokenization than on any single flow cycle.