News from manufacturers

Computex 2026: An Intelligent World Built on Silicon

Intel CEO Lip-Bu Tan delivered a keynote at Computex 2026, highlighting the company's innovations over the past five decades and the impact of x86 architecture. He also expressed gratitude to Taiwan for hosting the event.

Intel Announces New AI Innovations at Computex

At Computex 2026, Intel announced new AI innovations including rackscale AI infrastructure for scaling inference and agentic workloads based on Intel Xeon processors. The solutions are tailored to address specific industry challenges.

Related: Intel Xeon 6768P , Intel Xeon 6776P-B , Intel Xeon 6774P , Intel Xeon 6788P , Intel Xeon 6517P , Intel Xeon 6714P , Intel Xeon 6966P-C , Intel Xeon 6944P

Why Financial Institutions Are Converging on Transaction Foundation Models to Build Their Own Intelligence

Financial institutions are adopting transaction foundation models to unify AI systems previously siloed across fraud, credit, and risk tasks. These transformer-based models, trained on billions of financial events, provide contextual understanding of consumer behavior. NVIDIA's new developer example enables institutions to build such models on tabular transaction data.

“We move from weeks, or even in some cases months, in feature engineering to no time required for it at all” — NVIDIA

NVIDIA Jetson Brings Agentic AI to the Physical World

NVIDIA announced JetPack 7.2 and NemoClaw support for Jetson at COMPUTEX. JetPack 7.2 brings agentic AI skills, Yocto support, CUDA 13 on Jetson Orin, a performance boost to 241 TOPS on Jetson AGX Orin 32GB, and MIG support on Jetson Thor. NemoClaw enables agentic AI deployment on production-grade Jetson systems for robotics and industrial automation.

“Agentic AI is here, and Jetson’s programmability and high performance enable developers to instantly deploy physical AI agents in production at the edge.” — NVIDIA

NVIDIA AI Cloud Ecosystem Expands Worldwide to Meet Global AI Compute Demand

NVIDIA's AI Cloud ecosystem is expanding globally to meet demand for AI infrastructure from enterprises, startups, and governments. Partners are building AI factories using NVIDIA's full-stack AI infrastructure to support training, inference, and agentic AI applications across multiple regions.

NVIDIA Factory Operations Blueprint Gives Factories a New AI Brain

NVIDIA announced the Factory Operations Blueprint (FOX), a reference design for building autonomous factory manager agents that monitor real-time data and orchestrate specialized AI agents. The blueprint is optimized for NVIDIA DGX Station and integrates with industrial systems, automates model training, and manages factory workflows. Early adopters include Advantech, Foxconn, Pegatron, and Wistron.

“Foxconn projects an 80% improvement in root cause analysis time, a 15% increase in labor productivity and a 10% decrease in machine failure rates.” — NVIDIA

Taiwan’s Industry Titans Turbocharge World’s AI Infrastructure Buildout With NVIDIA

NVIDIA's Vera Rubin infrastructure involves over 500 ecosystem partners in Taiwan, with more than 1 million MGX rack components assembled across 25 factory sites. Partners like TSMC, Foxconn, and Quanta Cloud Technology are integrating NVIDIA's AI technologies into their own manufacturing processes, from chipmaking to factory operations, to improve efficiency and productivity.

“Foxconn estimates an 80% speed up in root-cause analysis time, a 15% increase in labor productivity and a 10% decrease in machine failure rates.” — NVIDIA

How Cosmos 3 Helps Physical AI Think Before It Acts

NVIDIA's Cosmos 3 platform aids physical AI systems in simulating and reasoning about actions before execution, enhancing decision-making in robotics and autonomous machines.

NVIDIA Levels Up Local AI Agents Across RTX PCs and DGX Spark

NVIDIA announced RTX Spark, a new class of Windows PCs for local AI agents, offering 1 petaflop of AI compute and 128GB of unified memory. The company also introduced DGX Station for Windows and updates to its local AI ecosystem, including OpenShell runtime integration with Microsoft security primitives. These products and updates are expected to arrive in fall 2025.

“RTX Spark’s 1 petaflop of AI compute and 128GB of unified memory can meet the computing demand of on-device agents” — NVIDIA

Related: NVIDIA GeForce RTX 5060 Ti , NVIDIA GeForce RTX 2060 Max-Q , NVIDIA GeForce RTX 3070 Ti , NVIDIA GeForce RTX 5080 , NVIDIA GeForce RTX 4070 Ti Super , NVIDIA GeForce RTX 2060 , NVIDIA GeForce RTX 5060 , NVIDIA GeForce RTX 4080

Intel Puts Agentic AI to Work with Xeon 6+, Networking, and AI Systems

Intel introduced Xeon 6+ processors featuring Efficient-cores, claiming industry-leading rack density for data center workloads. The company also expanded its 800 Series Ethernet portfolio with E835 controllers and adapters supporting up to 200GbE to reduce networking bottlenecks in AI, cloud, and edge environments.

Related: Intel Xeon 6768P , Intel Xeon 6776P-B , Intel Xeon 6774P , Intel Xeon 6788P , Intel Xeon 6517P , Intel Xeon 6714P , Intel Xeon 6966P-C , Intel Xeon 678X