Lam Research's Lam Capital recently held its fourth startup competition, drawing teams from the US, South Korea, Singapore, India, and Taiwan to vie for a sizeable prize pool. US startup Lightfinder won the top prize with a proposal centered on silicon photonics and intelligent software for a chip-scale spectrometer.
As the global semiconductor industry confronts the limits of Moore's Law, Huawei has unveiled a new roadmap aimed at extending chip performance growth through architecture, interconnect, and system-level optimisation rather than pure transistor miniaturisation.
Micron Technology said persistent supply constraints and accelerating AI demand are shaping its long-term technology roadmap, with 1-gamma DRAM, G9 NAND, and next-generation HBM products driving growth through 2027. The company is expanding EUV adoption, deepening collaboration with customers and foundry partners, and using multi-year supply agreements to support future capacity investments.
Cyient Semiconductors has secured a total of about US$30 million in financing from Edelweiss-managed funds and co-investors, comprising a US$10 million equity investment at a valuation of roughly US$500 million alongside structured debt. The deal bolsters its capital base to scale its power semiconductor and custom silicon offerings for global AI markets.
Nvidia's structural pivot to isolate its ACIE market and AMD's US$10 billion investment in Taiwan infrastructure signal a profound realignment in the AI chip war. Both developments reflect a shared urgency to expand beyond traditional hyperscale clouds into the booming, highly lucrative global enterprise, industrial, and sovereign AI factory frontiers.
Lam Research CEO Tim Archer said artificial intelligence and robotics can help chipmakers improve fab productivity as the semiconductor industry faces memory capacity constraints, chip-scaling limits, and growing demand for faster equipment delivery.
Nvidia's decision to sell its Vera CPU as a standalone chip could create a new source of demand for low-power DRAM, adding another pressure point to the already tight memory supply chain.
At its earnings conference, Weltrend Semiconductor described the first quarter of 2026 as an exceptional one. Despite the period being historically slow, the company posted year-over-year and sequential growth. Nearly all major product lines demonstrated strong growth momentum, with server-related products standing out in particular. Order visibility is now expected to remain strong throughout the full year. Meanwhile, AI servers are becoming increasingly diversified, with demand across GPUs, ASICs, and CPUs growing almost simultaneously, underpinning a highly promising operating outlook.
Nan Pao Resins Chemical is accelerating its push into the high-end semiconductor materials market through a joint venture with Advanced Echem Materials Company and Trusval Technology, forming Advanced Pao Trusval Technology to target advanced packaging adhesive materials.
Anthropic has reportedly approached Microsoft about renting AI computing power running on Microsoft's in-house chips to expand support for its Claude model business. The move is a positive sign for Microsoft and could generate momentum for the mass production of its recently unveiled Maia 200 chip, while ASIC players such as Global Unichip and Ethernet chip suppliers Marvell Technology and Broadcom also stand to benefit.
The shift toward 800V high-voltage direct current (HVDC) power architectures in AI data centers is driving a surge in demand for power semiconductors, boosting shipments for Taiwanese lead frame suppliers SDI Corporation and Jih Lin Technology and raising expectations for double-digit revenue growth in 2026.
More coverage
