Stories
Tracks the technical systems, bottlenecks, and standards that reshape industrial power before they become mainstream narratives.
The US DRAM market is undergoing a structural transformation driven by AI and high-performance computing. As AI model sizes expand, the demand for high-speed, high-capacity DRAM is surging, spurring the adoption of new technologies like HBM. At the same time, the US is strengthening domestic DRAM supply chain construction through policies like the CHIPS Act to address geopolitical risks.
Alan D. Thompson proposed a conservative definition of AGI, emphasizing that AGI should possess the ability to act in the physical world. He pointed out that although current LLMs have made progress in cognitive abilities, there is still a significant gap in embodiment. The article reviews the definition of AGI, the necessity of physical embodiment, and lists key technologies and model milestones in AI development up to 2026 and before.
The U.S. Department of Commerce announced that, under the CHIPS and Science Act, it has signed letters of intent with nine companies, providing a total of $20.13 billion in federal incentives to support key research and development and manufacturing in the field of quantum computing, in order to solidify the United States' leadership in this cutting-edge technology.
This paper analyzes that against the backdrop of the rapid development of AI infrastructure, memory, especially High Bandwidth Memory (HBM), is no longer a simple matter of bit production, but rather a strategic bottleneck and opportunity that determines overall system performance and competitiveness. The scaling of AI will no longer be solely determined by computational power, but by the synergistic efficiency of the system, which includes "compute + memory + bandwidth + packaging + power consumption + cooling," where memory has become a key strategic constraint and opportunity.
According to a MarketsandMarkets report, the global semiconductor manufacturing equipment market is projected to grow from $166.35 billion in 2025 to $344.36 billion by 2032, at a compound annual growth rate (CAGR) of 11%. Demand for advanced-node chips driven by AI and high-performance computing is the primary growth driver, with the Asia-Pacific region holding the dominant share and the Americas region experiencing the fastest growth.
Global challenges and technological opportunities are reshaping materials science, as four young scholars blaze new trails with diverse perspectives and interdisciplinary approaches.