Recently, global risk assets have experienced significant declines. In his post-mortem report, Rich Privorotsky, a senior trader at Goldman Sachs, pointed out that this round of declines was not caused by a single factor, but rather a systemic sell-off process resulting from the interplay of multiple factors, their sequential transmission, and eventual escalation. The entire process began with a shift in monetary policy expectations, progressed through a divergence in sector narratives and a reversal in retail sentiment, and was ultimately triggered by quantitative funds leading to concentrated selling pressure, pushing the market from a technical correction towards a structural decline.
The starting point of this market turbulence lies in the subtle changes in Federal Reserve policy. Despite contradictory signals from employment data—steady job growth but an unemployment rate rising to 4.44%, coupled with continuous downward revisions to employment figures—which should have prompted the Fed to signal a more dovish stance, it maintained its hawkish position. Market expectations for a December rate cut have essentially fallen to zero. This divergence between policy and data became the first domino in the decline.
The deeper reason for the tech stock adjustment lies in an internal restructuring of the artificial intelligence investment thesis. Although NVIDIA (NVDA) reported strong earnings, it is no longer the sole focus of AI investment. The real turning point was the breakthrough progress of Google’s (GOOGL) Gemini-3 model. This “disruptive model” is reshaping the AI ecosystem, leading to delayed product cycles for other companies, increased capital expenditures, and greater uncertainty regarding returns on investment. The failure of enterprise software stocks like Oracle (ORCL) to follow NVIDIA’s gains reflects market concerns about a “winner-takes-all” landscape.
The sharp volatility in the cryptocurrency market served as a barometer for retail investor risk appetite. Retail investors, who had adhered to the “diamond hands” strategy of buying the dip over the past two years, actively cut positions this time, while whale accounts also engaged in consecutive large sell-offs. This panic quickly spread to unprofitable tech stocks and AI concept stocks, for instance, Palantir turning from gains to losses. This indicates a substantive downgrade in retail risk appetite and a fundamental change in market structure.
What truly pushed the market into a “stall-speed decline” was the concentrated action of systematic funds. Since August, trend-following funds (CTAs) holding over $500 billion in long positions triggered consecutive sell orders after indices broke through key technical levels.
From a fundamental perspective, AI investment is facing the challenge of rising capital costs. The corporate bond market is about to see a wave of issuance, and the construction of AI data centers is highly reliant on debt financing. If credit spreads have bottomed, the attractiveness of carry trades under tightening spread conditions will diminish. Capital is no longer cheap, which may force a slowdown in the pace of AI expansion—a risk the market had not fully priced in previously.
The Goldman Sachs trader maintains the view that the S&P 500 mini futures could test the 6500 point level and notes that market stabilization requires three conditions: CTA positions being cleared out, excessive retail long positions being squeezed, and at least two of the following three being achieved: cryptocurrency market stabilization, a clear dovish pivot from the Fed, and supported AI capital expenditures. Although the market may be oversold in the short term, the fundamental value of AI technology remains unchanged. The true long-term winners will be those labor-intensive enterprises that achieve marginal profit expansion through automation.
