--- title: "Trump’s return to office is not far off. Risk appetite can start to recover." type: "Topics" locale: "en" url: "https://longbridge.com/en/topics/39867782.md" description: "By end-Q1 2026, risk appetite has shifted sharply from the AI asset frenzy of 2025, driven by early-year moves by Trump, rapid progress in AI, and changes in USD liquidity. But how much of this is a short-term repricing of war risk versus a deliberate realignment with long-term trends?After a rapid style rotation, what are the investable opportunities in 2026? In this strategy weekly, Dolphin Research offers a brief take.1) 2026: What is the ultimate trade? As the year began, the market had yet to come down from the AI leaderboard race..." datetime: "2026-04-13T13:03:32.000Z" locales: - [en](https://longbridge.com/en/topics/39867782.md) - [zh-CN](https://longbridge.com/zh-CN/topics/39867782.md) - [zh-HK](https://longbridge.com/zh-HK/topics/39867782.md) author: "[Dolphin Research](https://longbridge.com/en/news/dolphin.md)" --- # Trump’s return to office is not far off. Risk appetite can start to recover. By the end of Q1 2026, risk appetite has shifted sharply from the collective AI euphoria of 2025. A string of headline-grabbing moves by Trump, the breakneck pace of AI, and swings in USD liquidity have all pushed investors to recalibrate. How much of this is a short-term swing in war-related risk appetite versus a deliberate alignment with longer-term trends? After a rapid style rotation, what exactly are the investable themes in 2026? Dolphin Research offers a brief take in this weekly strategy note. **1) 2026: What Is the Ultimate Trade?** As 2026 began, markets had yet to step back from the AI leaderboard race. The launch of OpenClaw, the internalization of various skills into models, and more grounded productization drove deeper AI penetration into mainstream perception. Capital started to worry about the collapse of SaaS terminal value, triggering a broad selloff across software under doomsday narratives. The shift was swift and indiscriminate. At the same time, as Big Tech’s capex surged, pressure spread beyond industrial internet SaaS to consumer internet names. These companies are either footing today’s AI build-out bills or seen as likely ‘losers’ in the AI era. In Dolphin’s watchlist, the worst decliners are industrial internet SaaS, while the best performers are producers of capital goods and inputs benefiting from AI capex. On the consumer internet side, valuation cuts were broad-based. Giants faced short-term cash flow mismatches as capex rose, or were deemed to have business models ripe for AI disruption. Either way, multiples compressed. Two forces are now shaping market direction. Politically, the U.S. has escalated from a decapitation strike on Venezuela to rhetorical ‘epic fury’ toward Iran, and these moves ultimately point to energy, resources, and critical energy chokepoints. Layered on top is a severe mismatch in power infrastructure needed for AI build-out. With that backdrop, energy, metals, and other traditional resource plays have rallied even more than semiconductor capital-goods names. This differs from 2025’s straight-up contest over whose model was better, such as Google’s, or who could use AI to drive revenue and efficiency. That narrative sent PLTR and even META soaring. In 2026, amid macro uncertainty, the market is chasing what’s highly visible and near-term. Near term, who benefits with certainty from AI capex? Longer term, if AI truly replaces human labor, what demand is structurally and perpetually rewritten? These are the key questions anchoring flows. Near term, one clear answer is memory, where pricing has already adjusted. Up the chain, capex equipment makers and networking vendors are also in the slipstream. Over the long run, AI replacing labor is essentially compute replacing brainpower. Compute runs on electricity, whereas human cognition is powered by glucose. That shift to electricity plus compute as ‘silicon-based intelligence’ implies AI will fundamentally reshape future energy demand. The knock-on effects should not be underestimated. Energy and commodities may take time to transition, but the opportunity set looks long and persistent. We view this as a secular theme that unfolds over years. **Dolphin Research believes the re-rating in energy and other resource-linked assets will take a long time to play out. The market will need time to fully reprice these shifts.** **Between traditional internet and AI, whether the endgame is co-construction or full disruption remains uncertain. Who bears the risks of the infrastructure build and how business models are reshaped will also take time to verify, creating trades when internet assets are oversold.** **Therefore, for 2026, energy and commodity-linked assets look like ongoing opportunities. In the AI era, the struggle and symbiosis between traditional internet and AI also create sell-rally, buy-dip trading setups.** **2) TACO: Trade the Outcome, Not the Process** The Trump administration clearly recognizes energy’s centrality in AI’s endgame. As tariff skirmishes cooled, an energy land-grab began almost immediately. If we set aside the ebbs and flows of war and negotiations, and how long the tug-of-war lasts, the current U.S. macro picture is telling. The following points are most relevant. a) On the consumer side, household wealth growth relies excessively on asset appreciation. Persistent weakness in equities not only depresses returns but also erodes consumption willingness in the real economy. b) Sustained high oil prices lift inflation expectations. With the U.S. facing a heavy interest burden, relying on high oil to prop inflation and squeeze household affordability is not sustainable. This implies that regardless of the outcome in Iran, and whether hostilities shift from high-intensity to a prolonged, low-intensity stalemate, the administration will push to restore shipping through the Strait of Hormuz ahead of elections. A paper peace would help craft a narrative of having ‘won.’ In the U.S. electoral calendar, roughly Mar–Jun sees each party select its candidates. From Jun to Oct, the two parties then square off. If the war drags on, it clearly hurts the ruling party. Hence, de-escalation incentives are strong. Practically, from mid-Apr to Jun, we should see signs of de-escalation in the Strait of Hormuz. Volatility during negotiation back-and-forths can create buy-the-dip opportunities. This contrasts with the war’s onset, when excessive optimism saw risks build to the upside. Positioning should reflect that regime shift. **3) Portfolio Performance** Last week, Dolphin Research’s virtual portfolio Alpha Dolphin made no changes. The portfolio returned 2% for the week, underperforming CSI 300 (+4.4%), MSCI China (+3.1%), HSTECH (+3.9%), and the S&P 500 (+3.6%). Since inception on Mar 25, 2022 through last week, the portfolio has gained 115% in absolute terms. Excess return vs MSCI China is 97%. On an NAV basis, the initial $100 mn of virtual capital has grown to over $218 mn. Compounding has been steady. **4) Single-Name P&L Contribution** Last week, as the war moved into a more negotiable phase, most asset classes recovered. Within our holdings, high-quality AI exposures led the rebound. Given the portfolio has not been rebalanced for some time and the 2026 investment framework has shifted, we will refocus on active rebalancing in the near term. Expect more proactive management. **5) Asset Allocation** The Alpha Dolphin virtual portfolio holds 18 stocks and equity ETFs, with 7 at benchmark weight and the rest underweight. Outside equities, exposure is concentrated in gold, U.S. Treasuries, and USD cash, with equities vs gold/USTs/cash roughly 50:50. As of last week, asset allocation and equity position weights are as follows. See below for details. **6) This Week’s Key Events** As we move into mid-Apr, U.S. earnings season ramps up. Two semiconductor heavyweights, ASML (ASML) and TSMC (TSM), will report. Other key names include CATL and Netflix (NFLX), with focus points summarized below. We will track catalysts closely. **Risk Disclosure and Statement:**[**Dolphin Research Disclaimer and General Disclosure**](https://support.longbridge.global/topics/misc/dolphin-disclaimer) For recent weekly portfolio reports by Dolphin Research, please refer to: [‘The Most Grounded Way to Start: Dolphin’s Portfolio Is Off and Running’](https://longbridgeapp.com/topics/2764818?channel=t2764818&invite-code=032064) ### Related Stocks - [GOOG.US](https://longbridge.com/en/quote/GOOG.US.md) - [GOOGL.US](https://longbridge.com/en/quote/GOOGL.US.md) - [000300.CN](https://longbridge.com/en/quote/000300.CN.md) - [399300.CN](https://longbridge.com/en/quote/399300.CN.md) - [MSCI.US](https://longbridge.com/en/quote/MSCI.US.md) - [META.US](https://longbridge.com/en/quote/META.US.md) - [PLTR.US](https://longbridge.com/en/quote/PLTR.US.md) - [.SPX.US](https://longbridge.com/en/quote/.SPX.US.md) - [SPY.US](https://longbridge.com/en/quote/SPY.US.md) - [STECH.HK](https://longbridge.com/en/quote/STECH.HK.md) - [07552.HK](https://longbridge.com/en/quote/07552.HK.md) - [03032.HK](https://longbridge.com/en/quote/03032.HK.md) - [07226.HK](https://longbridge.com/en/quote/07226.HK.md) - [03033.HK](https://longbridge.com/en/quote/03033.HK.md) - [513130.CN](https://longbridge.com/en/quote/513130.CN.md) - [GGLS.US](https://longbridge.com/en/quote/GGLS.US.md) - [GGLL.US](https://longbridge.com/en/quote/GGLL.US.md) ## Comments (2) - **流水滔滔不绝 · 2026-04-13T13:22:03.000Z · 👍 1**: How to copy Dolphin Research's Alpha dolphin portfolio? 🤔 - **LiSOX.** (2026-04-13T17:01:52.000Z): Good question