AI TRENDS | Garrett Jin Says AI Supply-Chain Funds Are Shifting From Memory Chips to Hyperscalers

A market structure shift is underway this week, with capital within the AI industry chain being reallocated, according to Garrett Jin, an agent for the “1011 insider whale.” According to Odaily, Jin said the first change is that memory chips are showing signs of a near-term peak.

Jin said Micron’s share price met resistance around $1,250 and pulled back. He added that despite earnings beating expectations, the stock fell on heavy volume, which he described as a typical pattern of weakening after positive news is priced in.

He also said funds have been flowing out of the memory sector. Jin noted that DRAM-related ETFs declined on heavy volume, while South Korea’s SK Hynix and Samsung Electronics also weakened. He cited data showing that over the past two months, foreign investors have withdrawn more than 72.7 billion won from South Korea’s stock market.

Jin said the second change is a shift of funds toward AI hyperscale cloud providers rather than small- and mid-cap AI-themed stocks, naming Google, Microsoft, and Amazon as key beneficiaries.

He said that when the chip sector came under pressure last Friday, GOOG and MSFT stabilized on heavy volume, and that META’s rise on heavy volume this week reinforced the trend.

Jin attributed the rotation to what he called a “token optimization trend,” arguing that as more simple tasks are handled by lower-cost models, value will increasingly concentrate in token-billed cloud services and orchestration layers rather than the base model layer. He said this supports a strategy of focusing on potential catch-up gains in hyperscalers.