Sandisk drops 26% amid fears over Google's TurboQuant compression algorithm, but with Q2 data center sales surging 64% quarter-over-quarter, margins projected at 65-67%, and a $671 price target suggesting 17-18% undervaluation, analysts say the sell-off is overblown and the stock presents a strong buying opportunity.
Google launches Gemma 4, a new open-weight AI model family with four sizes optimized for both large-scale and mobile use, while switching to the developer-friendly Apache 2.0 license and teasing a next-gen Gemini Nano 4 for smartphones ahead of Google I/O.
Cognichip raises $93M to revolutionize chip design with deep learning technology that promises to slash development costs by 75% and cut timelines in half, attracting investment from Intel CEO Lip-Bu Tan despite having yet to publicly demonstrate a finished chip.
Oracle is cutting thousands of jobs amid a 25% stock drop and mounting debt, as the tech giant bets big on AI infrastructure with plans to raise $50 billion and honor a $300 billion OpenAI deal, while analysts say slashing up to 30,000 employees could unlock $10 billion in free …
DeepSeek suffers its longest outage in over a year, with two incidents totaling more than eight hours of downtime on March 29-30, 2026, just as the Chinese AI startup prepares to launch its next major model, V4, a multimodal system trained with help from Huawei and Cambricon.
South Korean AI chip startup Rebellions hits a $2.34 billion valuation after securing $400 million in a pre-IPO round led by Mirae Asset Financial Group, while simultaneously launching two new AI infrastructure platforms, RebelRack and RebelPOD, as the company aggressively expands into the U.S., Japan, Saudi Arabia, and Taiwan.
French AI startup Mistral secures $830 million in debt financing to build a massive Paris-area data center powered by 13,800 Nvidia GB300 GPUs, set to go online in Q2 2026 as Europe races to close the AI infrastructure gap with U.S. giants like OpenAI and Anthropic.
Starcloud raises $170M at a $1.1B valuation to build AI data centers in orbit, already operating a satellite with an Nvidia H100 GPU, though major technical and cost hurdles — including chip cooling, power generation, and reliance on frequent SpaceX Starship launches not expected until 2028 — still stand between …
AI inference costs are set to plummet over 90% by 2030, but total spending will surge as agentic AI models consume dramatically more tokens per task, forcing companies to strategically route workloads or risk falling into a costly commoditization trap.
Google Research's new TurboQuant algorithm is revolutionizing AI efficiency, slashing large language model memory usage by 6x and boosting performance 8x without compromising output quality, using a two-step compression process that works on existing models with no additional training required.