Press Enter to search · ESC to close

US Stocks

Hedge Funds Bought Tech Stocks at Fastest Pace in 15 Months — Right Before an AI Policy Shock

Goldman Sachs prime brokerage data show hedge funds net bought US TMT stocks in 10 of the past 11 sessions, with long buying at the 97th percentile of the past five years. A weekend clash over AI development policy sent SK Hynix down 6% and SoftBank down 11%, threatening freshly rebuilt tech longs ahead of a Fed decision and a 20-year Treasury auction.

Hedge Funds Rebuild Tech Longs at a Five-Year Extreme

Hedge funds have spent the past two weeks aggressively rebuilding long positions in US technology, media and telecom (TMT) stocks — and they did it at the fastest clip in roughly 15 months, just as a fresh dispute over the direction of artificial intelligence development erupted over the weekend.

Prime brokerage data compiled by Goldman Sachs show net buying of US TMT equities in 10 of the past 11 trading sessions. Over the trailing two weeks, long buying has run at the 97th percentile of the past five years and marked the fastest pace since June 2025. Net purchases were broad-based, with semiconductors and semiconductor equipment, interactive media and services, and IT services leading the way. Information technology and communication services ranked as the two most heavily bought US sectors, and both flows were dominated by long additions rather than short covering.

Leverage Ticks Up, Macro Hedges Go On

The positioning shift came with a subtle change in leverage structure. Gross leverage among US long/short funds slipped 1.6 percentage points to 206.9%, sitting in the 20th percentile of the past year, while net leverage rose 1.7 points to 50.2% — only the 6th percentile of the trailing twelve months. The fundamental long/short ratio by market value climbed 2.2% to 1.64, the 26th percentile for the year.

At the same time, funds sold macro products — combined index and ETF positions typically used to hedge single-name exposure — at the largest scale since the week of April 3 last year. Short sales outpaced long purchases by a ratio of 3.2 to 1, more than two standard deviations away from the one-year average. Short interest in US-listed ETFs jumped 7.2% in a single week, the biggest weekly increase in six months, concentrated in credit and small-cap equity funds. Even after that surge, aggregate macro short exposure remains below this year’s peak, which coincided with the momentum trade topping out in June.

The AI Fight Arrives at the Worst Possible Moment

The catalyst for the reversal in sentiment was a policy clash over AI development. Anthropic chief executive Dario Amodei called for slowing the pace of frontier model development and urged government intervention to constrain open-source competitors. President Trump declined to go along with that approach, leaving the regulatory outlook for AI sharply uncertain.

Asian technology shares bore the first brunt. On Monday, SK Hynix fell more than 6% and SoftBank dropped 11%. Goldman Sachs trader Lee Coppersmith put it bluntly in his weekly note: taken together, the weekend delivered new risk to freshly rebuilt tech long positions at the worst possible time.

The timing is especially awkward because the sector’s long base had only just been reconstructed. Any prolonged regulatory uncertainty around model development, open-source licensing or export controls could force funds to trim exposure they accumulated over the past fortnight.

What to Watch Next

  • FOMC decision: Markets currently price roughly a 90% probability of a September rate hike, making the Fed the dominant macro event of the week. Last week’s stronger-than-expected core CPI was read as clearing the way for tightening.
  • Other central banks: The Bank of England reports Thursday and the Bank of Japan Friday.
  • Treasury supply: A $13 billion 20-year auction on Tuesday will test demand at the long end and could move duration-sensitive tech valuations.
  • AI policy headlines: Any further escalation between the White House and AI labs would hit semiconductor and platform names hardest, given how concentrated the new long positioning is.

Key Takeaways for Investors

The setup is a classic crowded-positioning risk. Hedge funds have just rebuilt tech exposure at a near-record pace while simultaneously buying macro hedges — a sign that conviction in single names is high but confidence in the broader market is not. That combination leaves the trade vulnerable to any negative catalyst, and AI regulation has now become one.

Investors should watch whether the Asian selloff spreads to US semiconductors at the open, whether ETF short interest in credit and small caps keeps climbing, and whether the Fed delivers a hawkish hike that compresses long-duration equity multiples. If policy uncertainty persists, expect funds to rotate out of high-beta AI names and into defensives — or simply to cut gross exposure further.

View original

Share
Risk notice This site provides news and information on the crypto, blockchain and Web3 industry for reference only and does not constitute investment advice or any promise of returns. Virtual currency-related activities are illegal financial activities in mainland China; digital asset prices are highly volatile; use at your own risk. This site does not provide trading, token issuance or related referral services.

Related Reading

Latest News

TREE NEWS share card
Long-press image above → Save to Photos / Share
Pitch us Feedback