Two critical flaws in PHP, the language behind an estimated 70% of websites, triggered a global patching scramble this week. Researchers using OpenAI models, working with security firms Wiz and Calif, showed the bugs could let attackers take over web servers without authentication and read memory that may hold passwords. The companies nicknamed the bug set PHP2Shell. One flaw had been in the language for 14 years. The US National Security Agency and the EU cybersecurity agency urged immediate patching and said unhardenable systems should go offline. Some European government sites did so. Cloudflare reported early attack attempts before the patch was public.
Intel has reached a construction milestone at its $28 billion Ohio chip campus, with the first steel truss lifted into place at the Ohio One site built by Bechtel. The project is at least five years behind its original timeline, and the company now points to operations in the 2030-2031 window. Intel says its next Arizona fab should start running by the end of 2028, while its most advanced 18A chips are already in high-volume production in Arizona. Analysts say the foundry business still needs a large outside customer, and Intel reports third-quarter results on October 29. The campus was announced in 2022 as the centrepiece of Intel's push to build leading-edge capacity in the United States.
At GTC Berlin, Nvidia showed the DSX POD, a reference design for building a complete Vera Rubin AI system inside a shipping container, and pushed co-packaged optics for denser networking between racks. On software it introduced the OpenShell reference architecture for running AI agents with guardrails. Google Cloud said A5X instances on Vera Rubin NVL72 systems are now generally available for European customers, and WPP demonstrated a creative agent built on Nvidia Nemotron models and OpenShell. The company also announced a compute donation worth $250 million over five years for UK research on agentic AI safety. Power efficiency was a central theme.
The Department of Homeland Security has ordered federal civilian agencies to remove all products and services linked to Chinese AI developer DeepSeek from their IT systems and to stop buying them. A notice filed on Monday and published in the Federal Register covers the Treasury, State and Justice departments but exempts the Defense Department. DHS said the models present multiple threat vectors, including the risk that sensitive data could be transferred to China. The order relies on a 2018 federal supply-chain law.
Samsung will build the base dies for Nvidia's custom high-bandwidth memory, NVHBM, on TSMC's 3-nanometer process rather than its own foundry, according to the Seoul Economic Daily. Industry estimates put TSMC's 3nm yield above 90 percent against roughly 60 percent for Samsung's 2nm line. The design moves the memory controller onto the base die, freeing GPU space and delivering up to 30 percent more bandwidth and 15 percent less power than HBM4E. It would be the first time the two foundry rivals have worked together on HBM, and it comes as UBS projects a 10 percent global HBM shortfall next year.
Ricursive Intelligence, founded by Anna Goldie and Azalia Mirhoseini, is building AI systems that design chips, learn from each iteration and improve the next design. Speaking at TechCrunch Disrupt, the founders said demand for compute could rise 10,000-fold in five years and that the answer is more efficient silicon rather than more data centres. Their approach optimises whole chips instead of partitioning them into dozens of blocks, a method that produced the unexpected donut-shaped layouts of their earlier AlphaChip work. They argue that cutting design time from years to a week would make custom chips viable for space, AR, VR and biological computing.
The memory shortage that started in AI data centres has reached consumer shelves. Samsung, SK Hynix and Micron have shifted production towards high-bandwidth memory for AI accelerators, and 32GB DDR5 kits that sold for under $100 now trade above $400 in many markets. Gartner forecast DRAM prices would rise 47% in 2026, and SK Hynix says its HBM, DRAM and NAND capacity is essentially sold out for the year. Graphics card prices have also climbed as AI buyers take cards in bulk.
TSMC reported third-quarter revenue of US$46.72 billion, up 41.2 percent from a year earlier and above its own guidance, with net income of NT$763.06 billion, a sixth straight record quarter. High-performance computing, the bucket for AI accelerators, grew 22 percent from the previous quarter to 68 percent of revenue, while consumer electronics fell to 1 percent. Chief executive C.C. Wei said AI demand remains far above TSMC's ability to supply. The 2nm node is already 8 percent of wafer revenue, and the company guided fourth-quarter revenue to US$52.3-54.3 billion.
Microsoft introduced Azure Maia 300 at the OCP Global Summit in San Jose, its third-generation custom AI accelerator. The company says the chip delivers up to 40% more tokens per megawatt than leading GPU-based systems, and that a 96-accelerator rack offers seven times the rack-scale throughput of Maia 200. Production deployments begin in 2027, alongside multi-gigawatt Azure Maia buildouts. Microsoft now calls Maia the primary AI inference system for its own products and services, a shift from treating custom silicon as a supplement to Nvidia and AMD hardware.
Google launched a prototype satellite on a SpaceX Falcon 9 from Vandenberg on 1 October, carrying four of its own Tensor Processing Units into orbit. The mission is part of Project Suncatcher, a first step towards running data centres in space. The chips run a version of Google's Gemma model for about 15 minutes at a time because of heat limits, and the test will measure how they handle radiation and the stress of spaceflight. Google plans two more satellites in 2027 to test laser links. Google calls the mission a very minimal test. SpaceX and the startup Starcloud are pursuing similar orbital computing projects.