8 results on this page
Lightelligence is shifting from computing with light to connecting with light, signaling a more pragmatic focus on photonic interconnects. Despite strong investor enthusiasm following a 383.62% first-day stock surge in Hong Kong, the company faces concentrated customers, slow receivables, declining margins, and supply-chain risks.
China is increasingly deploying supernodes—clusters of dozens or hundreds of chips and supporting hardware—to scale AI computing. By combining larger quantities of domestic chips, these systems aim to deliver competitive performance despite US export restrictions on advanced AI chips.
AI chip startup Fractile is reportedly in advanced talks to secure a valuation above $6.5 billion. The discussions follow a supply agreement with Anthropic and would represent more than a sixfold increase from its May valuation.
Marvell Technology has granted Google the right to purchase up to $12.2 billion of Marvell shares as part of a custom-chip arrangement. Google will buy Marvell’s custom chips, and Marvell’s stock rose as much as 14% after the disclosure.
Samsung reportedly increased prices for new orders using its 4nm, 5nm, and 8nm foundry processes in July. Price increases reportedly reach 15% for customers in China as AI demand fills available 4nm capacity.
ByteDance and Tencent have reportedly each received 10,000 NVIDIA H200 chips. The deliveries could expand Chinese companies’ access to high-end AI computing despite ongoing U.S. export restrictions.
Baidu’s AI-related revenue remained around half of general business revenue in the second quarter, but fell to 125 billion yuan from 136 billion yuan in the prior quarter. The company’s AI portfolio spans cloud infrastructure, applications, search, Kunlunxin chips, and Apollo, yet model competitiveness, monetization, and growth remain unresolved.
Super Micro Computer completed an independent investigation into its export compliance program. Investigators concluded that senior management had no knowledge of an alleged scheme to divert Nvidia chips.