○ The "2026 AI Index" by the Stanford Institute for Human-Centered AI (Stanford HAI) highlights that while AI experiences accelerated quantitative and qualitative growth, the global competitive landscape is restructuring from a U.S.–China duopoly into a multipolar system.

• AI model performance on the SWE-bench coding benchmark leaped from 60% to 100% in just one year, enterprise adoption reached 88%, and generative AI population penetration reached 53%, demonstrating a faster diffusion rate than the PC or the Internet.

• The performance gap between U.S. and Chinese models was virtually erased to 2.7 percentage points as of March 2026, while global corporate AI investments surged 130% year-over-year to USD 581.7 billion.

• South Korea entered the global top 3 with 5 "notable AI models," trailing behind the U.S. (50) and China (30), ranked 1st for the second consecutive year with 14.31 AI patents per capita, and recorded the world’s highest growth rate in AI adoption (+4.8 %p).

○ Despite strong quantitative metrics, South Korea faces accumulating structural limitations in talent, computing, power supply, and governance, which act as major obstacles to achieving its "AI G3" objective.

• Among 38 OECD countries, Korea ranked 35th in net AI talent inflow (-0.36 per 10,000 people), with 11,000 Korean AI professionals residing abroad (16% of the total), reflecting an accelerating brain drain.

• Korea’s AI wage premium stands at 6%—significantly lower than the U.S. (25%) and the U.K./France (15%)—with seniority-based pay systems and compensation rigidity cited as structural constraints to talent retention.

• With the enforcement of the AI Basic Act in January 2026, Korea established the world's second comprehensive AI regulatory framework following the EU; however, further clarification is required regarding criteria for high-impact AI, detailed business operator obligations, and administrative guidelines.

○ For South Korea to leap forward as an AI G3 power, establishing an integrated strategy across "Infrastructure–Talent–Governance" and differentiating in "Physical AI"—applying AI to industrial sites such as factories and robotics—emerge as urgent imperatives.

• Public-private capital mobilization is taking full shape, led by the government’s pursuit of a KRW 100 trillion AI investment plan and Samsung Electronics’ investment to expand semiconductor manufacturing capacity in 2026.

• Leveraging its industrial base—such as a robot density of 1,012 units per 10,000 employees (ranking 1st globally)—Korea holds strong potential to secure "global leadership in industrial AI applications."

• Ultimate success will be determined over the next 3 to 4 years by securing computing self-reliance, establishing talent-return mechanisms, and bolstering the industrial absorption capacity for AI.