Subscribe to our newsletter
Get the latest updates delivered to your inbox every day, and stay up-to-date for free 🧠📈
Share
Linkedin
Twitter
Facebook
Whatsapp
Copy Link
A federal jury convicted former Google engineer Linwei Ding on 14 counts of economic espionage and theft of trade secrets. Between May 2022 and April 2023, he stole over 2,000 pages of confidential data about Google's AI infrastructure, including TPU and GPU specifications, while secretly founding AI startups in China. The case marks the first AI-related economic espionage conviction in U.S. history.
SK Hynix achieved a historic milestone by surpassing Samsung Electronics in annual operating profit for the first time in 2025, posting 47.2 trillion won compared to Samsung's 43.6 trillion won. The shift reflects SK Hynix's commanding lead in high-bandwidth memory (HBM) chips critical for AI applications, though intensifying competition from Samsung and Micron in the HBM4 market could reshape the landscape.
A US House committee chair alleges Nvidia provided extensive technical assistance to Chinese AI startup DeepSeek, helping achieve breakthrough training efficiency with just 2.788M H800 GPU hours. Documents suggest the AI models were later used by China's military, intensifying debate over US export controls and national security risks in the AI race with Beijing.
Samsung Electronics reported a record-breaking fourth quarter with operating profit surging over 200% year-over-year to 20.1 trillion won ($14.03 billion). The South Korean tech giant's memory business hit all-time highs as artificial intelligence server demand created a global memory chip shortage, pushing prices higher. The company plans to begin mass shipments of next-generation HBM4 chips to customers including Nvidia next month.
Nvidia is reportedly exploring Intel Foundry to manufacture portions of its 2028 Feynman AI chips, using Intel's 18A or 14A process nodes for I/O dies and EMIB packaging technology. The move signals a strategic shift to diversify supply chain beyond TSMC, driven by capacity constraints and political pressure to boost US manufacturing.
Scientists at Lawrence Berkeley National Laboratory have demonstrated that thermodynamic computing could generate AI images using one ten-billionth the energy of current tools like DALL-E and Midjourney. The breakthrough research shows promise for addressing the high energy consumption of generative AI, though significant hardware development challenges remain before the technology can rival existing models.
Intel launched XeSS 3 with multi-frame generation technology, enabling ultraportable laptops to achieve frame rates previously reserved for dedicated GPUs. Testing on the Asus Zenbook Duo showed Cyberpunk 2077 jumping from 67 FPS to 217 FPS with the AI-powered upscaling technology. The feature also works on MSI Claw handhelds, potentially reshaping the future of gaming laptops and portable gaming.
ASML has become Europe's most valuable company through its monopoly on extreme ultraviolet lithography systems that cost $250 million each. These massive machines are essential for manufacturing the most advanced semiconductors powering the AI boom, with TSMC using them to produce chips for Nvidia and other tech giants.
South Korean memory giant SK hynix is establishing a $10 billion AI solutions company in the U.S., restructuring its California-based Solidigm subsidiary in the process. The move comes as the chipmaker posts record quarterly profits, with operating income surging 137% year-over-year driven by explosive demand for high-bandwidth memory chips used in AI data centers.
China approved imports of Nvidia's H200 AI chips for ByteDance, Alibaba, and Tencent in January, totaling over 400,000 units. But nearly two months later, sales remain frozen as the US State Department pushes for tougher restrictions. Chinese customers aren't placing orders until licensing conditions become clear.
Dutch semiconductor equipment giant ASML posted record fourth-quarter bookings of €13.2 billion, more than double analyst expectations and previous quarter results. The surge reflects chipmakers' growing confidence in sustained AI-related demand, with customers accelerating capacity plans for advanced logic and memory manufacturing. ASML projects revenue between €34-39 billion for 2026.
Senior US and Taiwanese officials held their sixth round of high-level tech and AI talks, signing agreements on AI and semiconductor supply chain security. The US State Department praised Taiwan as a vital partner whose advanced manufacturing sector fuels the AI revolution, while discussions covered drones, critical minerals, and economic security cooperation amid growing geopolitical tensions.
The global memory shortage driven by AI demand shows no signs of slowing, with forecasts predicting continued DRAM constraints through 2027. Yet smartphones are expected to increase RAM by 16% as manufacturers prioritize on-device AI capabilities, potentially pushing device prices higher as memory costs surge and production struggles to keep pace.
Shanghai Iluvatar CoreX has revealed a four-generation GPU architecture roadmap aimed at competing with Nvidia's next-gen platforms. The Chinese chipmaker claims its current Tianshu architecture already outperforms Nvidia Hopper, while its 2027 Tianquan design targets surpassing Rubin. The announcement highlights China's aggressive push to develop domestic alternatives in AI chips amid ongoing trade restrictions.
The European Union faces a steep challenge in the global AI race, producing only three foundation models compared to 40 from the US and 15 from China. Despite launching ambitious initiatives like the €200 billion InvestAI scheme and plans for AI gigafactories, Europe struggles with funding gaps, talent drain, and infrastructure shortfalls that threaten its competitiveness in artificial intelligence.
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.