CICC: How to hedge against AI risks?

CICC released a research report stating that although the AI revolution is still in its early stages, the market has begun to worry about slowing growth among leading vendors and whether future returns can cover current large-scale investment. The bank remains optimistic about the return prospects of AI assets and recommends addressing recent market volatility through asset allocation, "buying insurance" for AI assets. In addition to defensive assets such as cash, bonds, and high-dividend stocks, the dollar depreciation trade can also help hedge AI asset risks, and increasing gold allocation may be an effective strategy for this trade. In addition, the allocation appeal of US Treasuries may also gradually increase, and close attention is recommended. CICC's main views are as follows: From the AI trade to the "AI dollar," the dollar has become the "dividing line" for global stock market divergence Against the backdrop of a restructuring international monetary order, the bank believes that "de-dollarization" remains the broad trend, and the dollar faces long-term depreciation pressure. However, since 2026, the dollar's movement has been strong in stages. In addition to inflationary pressure triggered by geopolitical conflicts, the AI rally may provide key support. On the one hand, cross-border transactions in the AI industrial chain, such as advanced chips, cloud services, and computing power, are largely settled in dollars, and industrial expansion may increase dollar demand in stages, creating the so-called "AI dollar" effect. On the other hand, AI investment strengthens the growth momentum of relevant economies, especially the United States, where nearly 30% of the increase in real GDP in 2026 is expected to come from AI. Economic growth resilience creates conditions for the Federal Reserve to raise rates, and rising US Treasury yields push the dollar higher. But for economies with lower AI benefits, tighter external financing conditions and capital outflow pressure brought by a stronger dollar will instead drag on economic and market performance. Under the influence of the above mechanisms, the dollar has become the dividing line for divergent global stock market performance in the AI revolution: since 2026, the US, Japanese, and South Korean stock markets, which are deeply involved in the AI industrial chain, have been positively correlated with the dollar, while the Indian stock market and the Hang Seng Index, which have relatively limited AI benefits, have been negatively correlated with the dollar. Chart 1: Since 2026, the Nikkei 225 has been positively correlated with the dollar Chart 2: Since 2026, South Korea's KOSPI has been positively correlated with the dollar Chart 3: Since 2026, the Hang Seng Index has been negatively correlated with the dollar Chart 4: Since 2026, the Indian stock market has been negatively correlated with the dollar Chart 5: Since 2026, correlation between the dollar index and major asset performance: the dollar has become the "dividing line" for global stock market divergence AI trade volatility has increased; in addition to increasing defensive assets, the dollar depreciation trade can also serve as a hedge Although the AI revolution is still in its early stages, the market has begun to worry about slowing growth among leading vendors and whether future returns can cover current large-scale investment. The market expects the year-on-year growth rate of capital expenditure by the US AI Big Seven to fall from about 90% in 2026 to 36% in 2027, and uncertainty over investment returns has also intensified volatility in AI assets. Chart 6: The market expects AI capital expenditure to continue growing in 2027, but at a markedly slower pace Against this backdrop, increasing allocation to defensive assets such as cash, Chinese bonds, and high-dividend stocks is reasonable. In addition, based on the analysis above, a cooling of the AI trade may weaken the US relative growth advantage and dollar demand, and the dollar depreciation trade is also expected to hedge adjustment pressure on AI assets. Another advantage of the dollar depreciation trade is that its profit scenarios are not limited to a weakening AI rally: if easing US-Iran tensions push oil prices lower, or if US inflation improves markedly and prompts the Federal Reserve to shift toward rate cuts, AI assets benefit from improved liquidity and the dollar depreciation trade benefits as well, while AI stocks may also rise simultaneously. Therefore, compared with simply reducing AI assets, positioning for the dollar depreciation trade can both retain exposure to AI upside and provide portfolio protection under multiple scenarios. The main risk to this hedging logic is a dollar liquidity crisis. If a sharp AI market adjustment triggers concentrated global asset selling and investors rush to liquidate and obtain dollar cash, the dollar may strengthen and move inversely to stocks, rendering the hedge ineffective. But the current AI market adjustment is mainly reflected in a decline in risk appetite and asset reallocation caused by concerns over investment returns, and has not yet evolved into a disorderly liquidity shock; the dollar and tech stocks are more likely to be positively correlated. During the recent adjustment in US tech stocks, the dollar index also fell from its October 14 high, consistent with this judgment. The above analysis mainly focuses on the short-term relationship between exchange rates and the stock market; over a longer horizon, even if the AI revolution succeeds, stock gains may not necessarily be accompanied by dollar appreciation, because the "reverse Balassa effect" brings real exchange rate depreciation pressure. Gold is an effective tool for positioning for the dollar depreciation trade, and increasing allocation is recommended The most direct way to position for the dollar depreciation trade is to short the dollar, but foreign exchange trading faces many restrictions, and increasing gold allocation may be a more convenient strategic expression. Theoretically, gold, as an alternative asset to the dollar system, has a basis for negative correlation with the dollar, but because gold prices are also affected by multiple factors such as interest rates, inflation, and safe-haven demand, this relationship has not always been stable historically. But entering 2026, the negative correlation between gold and the dollar has strengthened markedly, with the correlation rising to -81.2%. The dollar and gold have become two sides of the same coin, making increased gold allocation an effective tool for the dollar depreciation trade. Chart 7: Before 2026, the correlation between gold and the dollar was not stable Chart 8: Since 2026, the negative correlation between gold and the dollar has strengthened markedly, with a correlation coefficient as high as -81.2% By comparison, the impact of oil prices and US Treasury yields on gold is more reflected in disturbances: sharp rises and falls in oil prices during the year did not consistently dominate gold price movements, and rising US Treasury yields also failed to consistently suppress gold performance. Chart 9: Oil prices fell in May-June but did not benefit gold; oil prices surged in August but did not suppress gold Chart 10: After July, US Treasury yields accelerated upward but did not prevent gold from rising in August In addition to gold, US Treasuries are also a potential choice for positioning for the dollar depreciation trade, but the win rate may be lower than that of gold: if dollar depreciation stems from falling inflation and Fed rate cuts, both gold and US Treasuries may benefit; if dollar depreciation stems from rising US debt risk or shaken dollar credit, gold may benefit from demand for alternative assets, while US Treasuries may fall due to rising term premiums. Historical review also supports the above judgment. The bank reviewed all 91 dollar depreciation periods lasting at least 14 days since 2000; gold rose 84 times, a win rate of 92.3%, with a median period return of 5.46%; during the same periods, 2-year and 10-year US Treasury yields fell 68.1% and 64.8% of the time, respectively. Chart 11: During dollar index declines since 2000, gold's win rate was higher than that of other major asset classes In summary, the bank recommends that, in addition to increasing allocation to high-dividend stocks, cash, and Chinese bonds, investors increase gold allocation and position for the dollar depreciation trade to hedge AI asset risks. The allocation appeal of US Treasuries may also gradually increase, and close attention is recommended.