Tired of ads? Enjoy an ad-free experience by signing up.
👩‍🍳 How we use AI at Tech in Asia, thoughtfully and responsibly.
🧔‍♂️ A friendly human may check it before it goes live. More news here

SK Group exports to surpass $81.6b on chip demand

SK Group is set to surpass 120 trillion won (US$81.6 billion) in exports in 2025, driven by high demand for advanced memory chips from SK hynix.

The South Korean conglomerate reached 87.8 trillion won (US$59.59 billion) in exports by Q3, a 19% rise year-on-year, according to company figures.

SK hynix, known for supplying high-bandwidth memory (HBM) chips to Nvidia, contributed 65% of SK Group’s export total as of Q3, up from 54% last year.

South Korea’s Q3 exports hit US$185 billion, with advanced memory chips accounting for US$46.6 billion.

SK hynix paid 4.3 trillion won (US$2.92 billion) in corporate taxes in the first nine months of 2025, compared to 94 billion won a year earlier.

The company’s market capitalization rose to 379 trillion won (US$257.16 billion), making it the second-largest on the Kospi after Samsung Electronics.

🔗 Source: The Korea Herald

🧠 Food for thought

Implications, context, and why it matters.

  •  SK hynix has 62% of High Bandwidth Memory (HBM), yet pressure is rising as Micron (a US-based memory maker) passed Samsung in Q2 2025 with 21% versus 17% 1. It finished HBM4 development with 40% better power efficiency and 10 Gbps data rates 1.
  •  Samsung targets a rebound with HBM3E qualifications (customer validation testing) by chip plus cloud buyers, HBM4 volume in 2026, while Micron ships HBM4 samples at up to 11 gigabits per second (Gbps) and expects to sell out 2026 output with an annualized revenue run-rate (projected yearly revenue based on orders) of $8 billion 1.
  •  The HBM market is converging around three players, with buying centered on next-generation memory bins (performance grades) and packaging capacity (advanced chip-and-memory assembly capacity) 1.
  •  Independent liquid cooling providers (circulating coolant to remove heat from chips and servers) see revenue as AI pushes rack density to 30-100 kilowatts (kW), which exceeds air cooling limits 2.
  •  Direct-to-chip systems (coolant delivered via cold plates mounted on processors) remove 70%-80% of heat and pay back within 2-4 years despite higher upfront costs 2.
  •  The AI datacenter liquid cooling market will rise from $3.2 billion in 2025 to $7.2 billion by 2030, pushed by generative AI and machine learning needing heavy compute 3.
  •  Power infrastructure vendors benefit as 48% of data center developers cite grid connection delays as obstacles, which lifts demand for alternatives such as microgrids (localized power systems that can operate independently of the main grid) and renewable deployments 4.

Recent SK hynix developments

Stay ahead in Asia’s tech landscape

You've reached your 2 free content limit for the month. Sign up for free to read the full story.

🏄 For casual readers / 👶 Free

Basic

US$0

Free forever

Get instant access to this article and more every month

0 premium content

Unlimited news briefs

5

5 articles

Ad-free reading experience

Just US$0 per day

⌛Sign up in 20s. No payment details needed.

📖 For learners / 👍 Starter

Lite

US$4.92/month

Billed annually at US$59/year

Get instant access to this article and more every month

4

4 premium content

Unlimited news briefs & articles

Ad-free reading experience

Just US$0.17 per day

Cancel anytime

Our subscriber community includes professionals from these companies:

Stay updated on the go with our mobile app.

Get latest insights with smoother, more personalized experience through TIA mobile app.