Ideas
AI chip demand broadens through diversification.
At CES 2026, Nvidia's keynote and industry demonstrations showed AI moving from concept to real applications. The semiconductor industry, led by Nvidia, must diversify beyond data centers into autonomous driving, robotics, and other chip solutions. This expands the ecosystem and increases semiconductor content per device, supporting broad AI semiconductor demand. Semiconductors remain strong, though 2026 upside may be more limited than 2025.
Nvidia diversification expands long-term AI ecosystem.
Nvidia must diversify beyond data centers into autonomous driving, robotics, and other chip solutions to preserve its dominance. Alpamayo, Jetson-based small robots, and Vera Rubin with 5x better inference are concrete steps. Rubin's launch at the end of 2026 should lift physical AI levels and broaden Nvidia's ecosystem, though valuation is high.
Samsung earnings grow but upside is capped.
Samsung Electronics' 12-month forward valuation gap resembles late 2023 at the start of the AI cycle. Earnings are set to surge, but the stock may deliver more limited upside than in 2025, perhaps 50-60% if forward PER reaches about 15x and market cap around KRW 1,300tn. Still constructive on the semiconductor cycle but with a higher base and more limited multiple expansion.
Capex limits weigh on semiconductor equipment.
Memory makers cannot aggressively increase capex because demand visibility is uncertain and mobile bit growth is slowing. Conservative capex forecasts mean Korean semiconductor equipment/materials companies may underperform the memory-focused rally.
Micron growth and margins drive re-rating.
Micron's operating margin is now comparable to TSMC, and its growth rate could beat Nvidia next year. As memory demand rises, Micron's target price upgrades are faster than the market's largest tech names, so its valuation gap can narrow.
KOSPI upside supports market broadening.
If the KOSPI moves into the high 4,000s or 5,000, the market should broaden rapidly beyond large caps into individual stocks, similar to how M7 leadership eventually lifts the broader market. This supports the KOSPI and a broadening stock-picking environment.
Humanoid robots enter real-use phase.
CES 2026 shifted from 'wow' to 'now' with real demonstrations. Humanoid robots are needed where wheeled robots cannot operate, especially in factories and defense. The largest demand markets are automotive and defense, so humanoid robotics is a major investable theme.
Tesla Optimus leads humanoid robot tier.
Tesla's Optimus is viewed as a tier above other humanoid robots, while Unitree and EngineAI are more showy or defense-focused. Tesla is positioned as a leading humanoid robot player with large potential demand markets.
Korean robotics startups lag Chinese competition.
At CES 2026, Chinese robot companies dominated with many exhibitors and impressive demos, while Korean robot booths were empty and Korean startups showed limited differentiation. The Korean robotics startup space looks vulnerable unless tied to stronger platforms like Boston Dynamics.
AutoEver gains from robot data and MSCI.
Hyundai AutoEver is a key data and software solutions provider for Hyundai Motor Group's robot and mobility future. Robot and defense data cannot be sent offshore, so a group company must own the data and solution layer. AutoEver also has potential MSCI inclusion, which could add passive demand as its market cap grows; this gives it higher beta to the Boston Dynamics pipeline.
Hyundai Motor and Mobis upside limited.
The recent surge in Hyundai Motor and Hyundai Mobis reflects indirect Boston Dynamics pipeline value rather than near-term earnings, and their fundamental upside is limited. The speaker prefers other group names with more direct pipeline exposure.
Glovis offers cheaper Boston Dynamics stake exposure.
Hyundai Glovis holds a stake value in Boston Dynamics that is larger than Hyundai Mobis's, while its market cap is only about half of Mobis's. If the recent rally reflects Boston Dynamics pipeline value rather than auto earnings, Glovis should have more upside momentum.
Special vessel demand extends shipbuilding cycle.
Trump's Greenland ambitions, Golden Fleet naval expansion, and Venezuelan oil logistics all require ships. Arctic routes need icebreakers; naval modernization needs warships; oil exports need tankers. In 2026, special vessels such as LNG carriers, warships, and shuttle tankers may be over half of orders, extending the Korean shipbuilding cycle.
Samsung Heavy leads icebreaker special-vessel demand.
Samsung Heavy Industries has icebreaker construction capability and won a large Russian icebreaker order. As Arctic routes and Greenland-related logistics increase demand for icebreakers and special vessels, Samsung Heavy is a direct beneficiary.
China begins AI hardware infrastructure cycle.
China is entering an AI hardware cycle, as seen at CES 2026 and with AI inclusion in domestic indices. Its push into AI hardware will require infrastructure, especially components and electric power, creating investment opportunities in the Chinese AI hardware supply chain.
New form factors re-rate IT component makers.
New hardware form factors for robots, autonomous driving, and physical AI should lift the multiples of IT hardware component makers. In past new-form-factor cycles, Korean component names such as Samsung Electro-Mechanics and LG Innotek moved from below-average valuations to above-average and delivered more than double upside.
KEPCO dividend and PBR re-rating potential.
With electricity price risk largely passed and power moving into a normal management phase, KEPCO can improve earnings and pay dividends. Global utilities paying 3%+ dividends often trade above 2x PBR, while KEPCO trades around 0.6x PBR. Long-term valuation and dividend re-rating potential are attractive, helped by potential tariff hikes and shareholder return demands at the March meeting.
Reform lifts holding company and utility dividends.
The March shareholder meeting season and completed commercial-law reform should increase shareholder proposals, including dividend demands. Dividend payouts for Korean holding companies and utilities are likely to rise, supporting these sectors.
KOSDAQ supply-demand improves from late January.
The KOSDAQ has suffered from poor supply/demand, with institutions avoiding it and short selling concentrated there. From late January or early February, pension KPI inclusion and a national fund launch should improve flows. The KOSDAQ must be supported because Korean AI, bio, and new growth companies list there.
Large-cap Korean bio leads KOSDAQ rally.
Bio should lead the KOSDAQ rally. Within Korean bio, large caps are preferable to small caps: Samsung Biologics, Celltrion, and Alteogen. Even if the bio rally comes in the second half, large-cap bio should outperform, and KOSDAQ bio exposure can be accessed through ETFs.
This 3PRO TV (삼프로TV) video, published January 13, 2026,
features Lee Hyuk-jin
discussing SMH, NVDA, 005930.KS, Korean semiconductor equipment/materials, MU, ^KS11, Humanoid robotics, TSLA, Korean Robotics, 307950.KS, 005380.KS, 012330.KS, 086280.KS, Korean Shipbuilding, 010140.KS, China AI hardware, Korean IT hardware component makers, 009150.KS, 011070.KS, 015760.KS, Korean holding companies, Korean utilities, ^KQ11, 207940.KS, 068270.KS, 196170.KQ.
20 trade ideas extracted by AI with direction and confidence scoring.
Speakers:
Lee Hyuk-jin
· Tickers:
SMH,
NVDA,
005930.KS,
Korean semiconductor equipment/materials,
MU,
^KS11,
Humanoid robotics,
TSLA,
Korean Robotics,
307950.KS,
005380.KS,
012330.KS,
086280.KS,
Korean Shipbuilding,
010140.KS,
China AI hardware,
Korean IT hardware component makers,
009150.KS,
011070.KS,
015760.KS,
Korean holding companies,
Korean utilities,
^KQ11,
207940.KS,
068270.KS,
196170.KQ
