Huawei announced it will accelerate the release of its next-generation AI chips by three to nine months to meet rising global AI compute demand. This move aims to enhance Huawei’s competitive position in the AI hardware market amid intensifying industry competition....
News

Ex-Intel Engineers Launch High-Speed Interconnect to Compete with NVIDIA NVLink in AI Infrastructure
A startup founded by former Intel engineers has launched a next-generation high-speed interconnect technology designed to compete with NVIDIA’s NVLink in AI infrastructure. The company claims superior bandwidth and scalability tailored for large AI compute clusters, with pilot deployments planned fo...

Crusoe has achieved a $30.9 billion valuation following a March 2026 funding round aimed at expanding its AI infrastructure. The company’s hybrid model, which combines waste energy capture with AI data center operations, positions it as a key player in sustainable AI compute amid increasing market d...

The U.S. Senate voted down a bill on September 18, 2026, that would have imposed stricter regulations on data center expansion amid surging AI infrastructure demand. The legislation's defeat highlights ongoing tensions between supporting AI growth and ensuring energy sustainability and grid reliabil...

Microsoft announced its Distributed Skills architecture on September 19, 2026, as a replacement for the Microsoft Copilot Platform. This new modular approach aims to improve scalability, flexibility, and security in managing multi-agent AI systems, leveraging Azure's cloud infrastructure. The phased...

AI chip startup Rebellions and Tokyo-based ai& have partnered to deploy next-generation AI chips in Japan, starting with ai&’s Tokyo data center in Q2 2026. The collaboration aims to strengthen Japan's AI hardware ecosystem and reduce reliance on foreign suppliers amid growing domestic demand. This ...

Crusoe has secured a $30.9 billion valuation in its latest funding round, led by investors focused on technology and sustainability. The company’s modular data centers use stranded energy sources to provide sustainable AI compute power, positioning Crusoe as a key player in the rapidly evolving AI i...

Amazon Web Services has launched the EC2 T8i instances powered by custom Intel Granite Rapids CPUs, offering up to 30% improved price performance over the T3 generation. These burstable instances target AI inference and other workloads with variable CPU demands, featuring enhanced networking and sec...

Amazon Web Services and Nvidia will deploy 2 million next-generation GPUs globally from 2027 to 2028, significantly expanding AI computing capacity. This deployment, featuring Nvidia’s Blackwell Ultra, Rubin, and Rubin Ultra models, targets growing demand for agentic and physical AI workloads and ai...

Huawei unveiled its Lingqu UnifiedBus in September 2026, a memory-fabric interconnect technology that supports clusters of up to one million processors, surpassing Nvidia's NVLink 4 maximum of 576 GPUs. The technology aims to enable unprecedented AI compute scale and efficiency for large language mo...

Simplilearn and Carnegie Mellon University have launched a new executive education program starting September 2026 to train business leaders in large language models and multi-agent AI systems. The curriculum combines technical fundamentals with strategic and operational insights to address growing ...
Agentic AI frameworks, combined with strategic partnerships and hardware innovations, are driving a fundamental shift in enterprise AI deployment. Open, modular architectures like Microsoft's Orchard and vertical-specific solutions such as Guidewire's insurance platform enable scalable autonomous AI...
AI infrastructure investment is expanding beyond traditional hyperscalers, driven by diverse sector demand and innovation from specialized chip startups. This diversification reshapes supply chains, intensifies competition, and creates new strategic considerations for AI infrastructure professionals...
Supabase’s open-source Evals framework introduces a realistic and transparent approach to benchmarking AI coding agents by testing models like Claude Code, Codex, and OpenCode on authentic platform-specific tasks. This advancement addresses key limitations of prior benchmarks, fostering fair compari...
The escalating energy demands of hyperscale AI data centers require sustainable power solutions. Combining large-scale renewable energy, exemplified by a nearly 1 GW solar hub in Texas, with expanding nuclear power capacity, as seen in China’s recent reactor approvals, offers a balanced approach to ...
Nvidia’s Vera CPU, AMD’s balanced CPU-GPU strategy, and emerging East Asian and open-source hardware projects are reshaping AI infrastructure chip competition. This analysis explores how these developments disrupt Nvidia’s CUDA dominance, introduce heterogeneous compute architectures, and influence ...
Schneider Electric has introduced 2.5MW prefabricated power modules designed to meet the high-density power needs of AI data centers. These factory-built units aim to reduce installation times and imp...
OpenAI’s 40% price cut on its GPT-5.6 Sol model and Anthropic’s launch of the specialized Mythos 5 model highlight a growing competition in AI deployment costs. This battle reflects broader shifts in ...
Delos Data has raised over $100 million to develop specialized networking infrastructure aimed at supporting agentic AI inference workloads. This funding will accelerate product development focused on...
Liquid Compute has raised $15 million to develop a regulated AI infrastructure marketplace focused on compliance and security for regulated industries. The platform aims to enable compliant AI compute...
Fujitsu announced its MONAKA AI inference chips designed for air-cooled data centers, targeting volume production in 2027. These chips aim to reduce power and cooling requirements compared to GPUs, en...
The escalating energy demands of hyperscale AI data centers require sustainable power solutions. Combining large-scale renewable energy, exemplified by a nearly 1 GW solar hub in Texas, with expanding...
Digital Realty Deploys ServiceFabric® MCP Across 800+ Data Centers to Enhance AI Workload Management
Digital Realty has deployed its ServiceFabric® Managed Control Plane, an AI-native programmable platform, across more than 800 data centers globally. The platform enhances real-time AI workload manage...
This opinion piece argues that celebrating AI models like Ox Alpha without transparency about their creators and inner workings is reckless. It calls for mandatory openness to ensure accountability, t...
Cornelis raised $205 million to develop AI hardware and software alternatives to Nvidia’s GPUs, signaling growing investor interest in diversifying AI infrastructure. The funding will support Cornelis...
Samsung Ventures led a $250 million funding round in Dutch AI chip startup Euclyd in March 2026, aiming to accelerate the development of AI inference chips that compete with NVIDIA’s GPUs. This invest...















