Intel has announced a range of AI-focused initiatives at Computex 2026, including new rack-scale AI infrastructure, an enterprise inference cloud platform, and strategic partnerships across manufacturing, healthcare, robotics, and life sciences. The announcements build on Intel’s recently launched Xeon 6+ processor family, which was unveiled earlier this week as part of the company’s broader AI infrastructure strategy.
Intel, Foxconn, and SambaNova Partner on Rack-Scale AI Infrastructure
Intel, SambaNova, and Foxconn have announced plans to develop rack-scale AI infrastructure aimed at data centres, hyperscalers, and intelligence centre deployments.
The solution combines Intel Xeon processors with SambaNova’s SN-50 Reconfigurable Dataflow Units (RDUs) to deliver AI inference performance with improved power and cost efficiency. Foxconn will provide system integration capabilities for the platform and plans to manufacture CPU-dense variants designed for cost-optimised inference, data processing, and hybrid AI workloads.
Intel says the rise of agentic AI is shifting the balance of AI infrastructure, increasing the importance of CPUs for orchestration and workload management alongside dedicated accelerators.
Vector Core Compute Debuts Disaggregated AI Inference Cloud
Intel also joined SambaNova, Vista Equity Partners, and Cambium Capital to showcase Vector Core Compute, a new enterprise-focused inference cloud platform.
The platform uses Intel Xeon 6 processors for orchestration and execution, SambaNova SN40 RDUs for inference decoding, and NVIDIA Blackwell GPUs for prefill workloads. During the Computex demonstration, the companies showcased what they described as the first real-world deployment of a fully disaggregated inference architecture operating from a data centre in Los Angeles.
According to Intel, Together.ai has become the first commercial customer of the platform, while Vista Equity Partners has secured early access to the infrastructure for its portfolio companies.
Intel Expands AI Collaborations Across Multiple Industries
Beyond infrastructure, Intel announced several strategic collaborations focused on developing industry-specific AI solutions.
Foxconn is working with Intel on rack-scale AI deployments, systems integration, and custom silicon opportunities. Siemens has expanded its partnership with Intel to explore purpose-built silicon across industrial automation, manufacturing, robotics, edge computing, and high-performance computing applications.
Intel is also collaborating with Hitachi on foundry technologies and quantum computing initiatives. Echo Neurotechnologies is working with Intel to explore neuromorphic computing, neuro-AI, brain-computer interfaces, and speech neuroscience technologies.
Meanwhile, Greenstone Biosciences plans to utilise Intel processors, purpose-built silicon, and the Intel Health and Life Sciences AI Suite to accelerate drug discovery using stem cells, organoids, genomics, and AI-driven research.
Core Ultra Series 3 Adoption Continues to Grow
Intel also provided an update on its Core Ultra Series 3 processor family.
According to the company, Series 3 processors now power more than 325 consumer and commercial PC designs. Intel says more than 130 customers have selected the platform for edge AI and robotics deployments spanning manufacturing, retail, smart cities, and industrial applications.
The company also confirmed that Intel Arc G-series processors targeting handheld gaming devices will begin shipping this month, expanding Intel’s presence in the growing gaming handheld market.
Intel Pushes Beyond Chips With a Broader AI Infrastructure Strategy
While much of the AI industry remains focused on accelerators and model training, Intel’s latest Computex announcements highlight a broader strategy centred on AI deployment, inference, and industry-specific solutions.
The company’s focus on rack-scale infrastructure, disaggregated inference, and vertical AI partnerships suggests Intel sees the next phase of AI growth being driven not just by larger models, but by the systems, orchestration layers, and industry platforms required to deploy them efficiently at scale.
