Back to News
Market Impact: 0.35

Big Shift in AI Stock Trade Drives Hunt for New Stars in Asia

TSM
Artificial IntelligenceTechnology & InnovationEmerging MarketsInvestor Sentiment & PositioningMarket Technicals & FlowsCompany Fundamentals
Big Shift in AI Stock Trade Drives Hunt for New Stars in Asia

As the multi-year equity rally tied to the launch of OpenAI’s ChatGPT enters its fourth year, major regional beneficiaries such as Taiwan Semiconductor Manufacturing Co. and SK Hynix are losing momentum, prompting investors to hunt for new Asian AI winners. Market participants are rotating toward smaller, lesser-known technology names like MediaTek and Zhongji Innolight amid a technological shift and rising bubble concerns, signaling a reweighting of positions rather than a broad-market unwind.

Analysis

Market structure: The AI trade is rotating from large-cap foundry/DRAM beneficiaries (TSM, SK Hynix) to fabless chip designers and specialty suppliers (MediaTek, optical components). That shifts pricing power from capacity owners to IP/design and niche supply players — expect gross-margin dispersion to widen by 200–500bps across subsegments over 6–12 months as orders reallocate and customers favor differentiated AI inference/optical solutions. Risk assessment: Key tail risks are US export controls or China countermeasures that could cut TAM for Asian AI suppliers (low-probability, high-impact) and rapid destocking if end customers slow purchases; both could erase 20–40% of near-term revenue for small names. In days–weeks watch order guides and Taiwan export data; in quarters watch capex plans and foundry utilization which determine whether rotation is structural or a short bubble. Trade implications: Favor long exposure to high-ASIC-content, fabless designers and optical component makers with revenue visibility and customer concentration limits; hedge with protective puts on large-cap foundries. Use relative-value pairs (long MediaTek, short TSM) to isolate secular design wins vs. cyclical capacity; allocate 1–3% per idea and layer by catalyst (earnings, order book updates) over 3–6 months. Contrarian angles: Consensus may over-rotate into small names that lack scale — many lack durable moats and have lumpy revenue, so volatility will be higher; conversely TSM may be oversold given structural foundry secular demand beyond AI (automotive, Apple). Look for companies where earnings revision momentum is just beginning (price-to-forward-revisions divergence >25%) rather than names already up 50–100% YTD.