Dutch lithography giant ASML has begun shipping its monumental $400 million High-NA EUV machines to Intel, a massive 150-ton bet designed to push the physical limits of chipmaking,. By enabling the production of transistor features as tiny as eight nanometers, this behemoth of a machine is the undisputed linchpin of the global semiconductor supply chain and will dictate the pace of AI hardware scaling for the next decade,.
Tencent has quietly rolled out its first native AI assistant, “Xiaowei” (小微), placing a highly visible green-eyed AI entry point directly on the WeChat homepage. This is arguably WeChat’s most significant update in years, as Xiaowei is deeply integrated into the app’s ecosystem, allowing users to execute tasks like searching their Moments timelines, summarizing chat histories, and directly calling mini-programs. By utilizing a user’s private social graph and chat memories as a unique context window, WeChat is leveraging its proprietary WeLM model to create a deeply personalized AI that general-purpose third-party chatbots simply cannot replicate.
Micron Technology Inc.’s blowout earnings report has reignited the artificial intelligence trade, sparking a rally across US equity futures and semiconductor stocks following a brutal $1.3 trillion two-day tech rout. The insatiable demand for AI infrastructure is driving unprecedented capital market activity, highlighted by SK Hynix Inc.’s pursuit of a landmark $29.4 billion US listing to fund its memory chip expansion.
Micron Technology shares surged over 10% in extended trading after delivering a massive fiscal third-quarter beat, driven by explosive AI-related memory demand. The company’s staggering 84.9% gross margin has pushed it past Nvidia and Meta, crowning Micron as the tech sector’s new margin king amid persistent global supply constraints.
Treasury Secretary Scott Bessent joined “Squawk Box” to outline the administration’s economic vision, stating that U.S. GDP growth can return to 3% before the end of the year. Bessent also assured investors that the U.S. Treasury will oversee frozen Iranian funds upon their release as part of the interim peace deal, mandating their use for U.S. agricultural and pharmaceutical goods. In real estate, Rick Santelli broke down the May new home sales decline as the housing market evaluates persistent affordability issues. The energy market saw Brent crude fall back to pre-war levels as tankers resumed exiting the Strait of Hormuz, prompting Energy Secretary Chris Wright to assert the U.S. has effectively ended Iran’s ability to close the vital shipping lane.
OpenAI and Broadcom have unveiled “Jalapeño,” OpenAI’s first custom-designed AI inference chip. According to a cnbeta report, the ASIC took only nine months from design to tape-out, a record-breaking pace aided by OpenAI’s own models participating in the layout and circuit design. A separate deep dive notes that the chip will be deployed by late 2026, marking a major step towards OpenAI’s goal of becoming a full-stack AI company capable of driving down inference costs for its advanced models and upcoming AI agents.
The flood of AI-generated code is reaching a breaking point on GitHub, prompting structural changes to how open-source maintenance works. After repositories like OpenClaw saw weekly pull requests jump from two to 3,400—mostly comprised of low-effort AI “slop”—GitHub officially introduced persistent PR limits to give maintainers a fighting chance to triage the noise. The shift signals a new era where maintaining open-source projects requires automated governance and reputation management just to filter out the automated garbage.
Bosnia and Herzegovina secured a 3-1 victory against Qatar in Wednesday’s FIFA World Cup match at Seattle Stadium, drawing tens of thousands of international fans to the city. The win likely sets up Bosnia to face the United States Men’s National Team in Seattle for the Round of 32 on July 1.
Microsoft’s Talos pipeline consciously traded maximum algorithmic recall for extreme specificity—surfacing just 1.3 candidate genomic variants per patient—to respect the severe operational bottleneck of human expert review time. This highlights a crucial architectural principle for deploying AI at scale: optimizing models for peak theoretical accuracy is counterproductive if the resulting false-positive rate overwhelms the human-in-the-loop workflow.
OpenAI and Broadcom have officially unveiled “Jalapeño,” a custom application-specific integrated circuit (ASIC) built to efficiently handle AI inference tasks for large language models. The custom hardware signals a massive strategic shift as OpenAI moves to build its own full-stack infrastructure, aiming to slash operational costs by up to 50%.
Meta emergency pauses employee AI-training tracking over massive data leak Meta has indefinitely suspended a controversial internal initiative that tracked employee computer usage to train AI models, following a severe data leak. The “Model Compatibility Initiative” (MCI) unintentionally exposed 45,000 internal databases—including private employee chats, prompt transcripts, and sensitive performance data—to the entire company. The breach has cratered internal morale, prompting fierce employee backlash against leadership’s authoritarian approach to AI data collection.