Latest AI News

SK Hynix forecasts the AI memory market to grow 30% annually through 2030, driven by hyperscale AI infrastructure investments. High-bandwidth memory (HBM) remains central to AI performance, with the company expanding in the US and advancing its product roadmap. Despite potential constraints like energy supply limitations, SK Hynix sees this growth rate as a conservative estimate, highlighting AI memory as a key structural growth area in semiconductors.

The U.S. now requires Nvidia and AMD to pay a 15% fee on revenues from sales of specific AI chips to China, including Nvidia's H20 and AMD's MI308. This unprecedented export control condition could generate billions in fees annually, potentially affecting pricing and availability in China's AI market. Both companies must comply with these rules while Chinese buyers may face higher costs and seek alternative solutions.

SK Hynix forecasts 30% annual growth in the AI memory market through 2030, driven by surging demand for high-bandwidth memory (HBM) essential for AI models like ChatGPT. The company dominates with 70% market share in HBM and is investing billions in new facilities while developing next-generation HBM4 technology. Despite potential pricing pressures as competitors like Samsung enter the market, SK Hynix's technological leadership and partnerships with NVIDIA position it to capitalize on the AI boom, with the market expected to grow from $15.1 billion to over $45 billion by 2030.

Chinese state media has raised national security concerns about Nvidia's H20 AI chips, with authorities claiming they could contain "backdoor safety risks" allowing remote tracking and disabling. The Chinese Cyberspace Administration summoned Nvidia executives amid accusations that the technology threatens cybersecurity. Nvidia firmly denies these allegations. The controversy unfolds against a backdrop of US-China tech rivalry, with Chinese officials suggesting they might promote domestic alternatives like Huawei's chips if security issues are confirmed. The situation highlights growing tensions in the global AI hardware market and supply chain vulnerabilities amid geopolitical pressures.