AI Agent Competitive
Intelligence
Intelligence Journeys
AI Use Cases for Utilities
Start New Project →
Private Broadband for Utilities
Start New Project →

Amazon

AI Capex Race: Amazon and Google Lead

AI Capex Race: Amazon and Google Lead
Amazon and Google currently lead the AI capex race, with Microsoft and Meta not far behind, and the prize is control over scarce compute, power, and network resources that define the next decade of cloud and AI services. For telecom and infrastructure players, the opportunity is immediate: deliver power-adjacent, fiber-rich, AI-ready capacity with speed and predictable SLAs. For enterprises, the mandate is pragmatic: secure capacity, design for portability across heterogeneous silicon, and enforce cost governance as inference scales. The winners will be those who pair aggressive buildouts with disciplined execution—turning record capex into durable platforms and customer outcomes.

AT&T–AWS Connectivity Deal: LEO Satellite, Fiber, and Cloud Adjacency

AT&T–AWS Connectivity Deal: LEO Satellite, Fiber, and Cloud Adjacency
AT&T is deepening ties with Amazon by pairing its national fiber assets with AWS cloud and AI tooling while adding low Earth orbit connectivity from Amazon’s satellite network to fill coverage gaps for business customers. The collaboration has two pillars: cloud modernization on AWS and satellite-enabled reach via Amazon’s LEO network, with AT&T also supplying fiber capacity into AWS data centers to bolster high-performance infrastructure. Amazon’s LEO constellation will deliver fixed broadband connectivity for AT&T Business customers in areas where terrestrial options are limited, enabling primary service in hard-to-reach sites and resilient backup for SD‑WAN architectures.

Nvidia OpenAI investment: Huang rejects $100B stall

Nvidia OpenAI investment: Huang rejects $100B stall
Nvidia’s CEO is publicly reaffirming confidence in OpenAI even as reports suggest the companies may narrow the scope of an ambitious, nonbinding plan announced last fall. During a visit to Taipei, Nvidia CEO Jensen Huang dismissed talk of friction with OpenAI and said Nvidia will participate in OpenAI’s next funding round. Recent reporting suggested Nvidia has emphasized the nonbinding nature of its plan to invest up to $100 billion and build roughly 10 GW of compute for OpenAI, and that both parties are re-examining scope and terms.

India’s Zero-Tax AI and Cloud Exports Through 2047

India’s Zero-Tax AI and Cloud Exports Through 2047
New Delhi has unveiled a sweeping tax holiday to capture the next wave of AI and cloud build-outs, positioning India as a long-term base for exporting compute. Foreign providers that deliver cloud and data center services to customers outside India will pay zero corporate tax on those revenues through 2047, provided workloads run from facilities in India. The budget also introduces a 15% cost-plus safe harbor for Indian data center units serving related foreign parties, simplifying transfer pricing for global delivery hubs. For cloud providers, it strengthens the business case to place GPU clusters, storage, and interconnect in India to serve overseas demand, not just local workloads.

NTT DATA and AWS Scale AI and Cloud in APAC

NTT DATA and AWS Scale AI and Cloud in APAC
NTT DATA and AWS have signed a multi-year strategic collaboration aimed at accelerating cloud modernization and responsible agentic AI adoption, with clear implications for APAC enterprises and telecoms. The agreement expands joint go-to-market and delivery across four pillars: AI-driven cloud transformation, industry cloud solutions, AI-enabled managed services and customer experience, and sovereign cloud for regulated workloads. NTT DATA has created a dedicated AWS Business Group with close to 11,000 AWS-certified experts and plans to certify nearly 10,000 more in three years. APAC boards want measurable AI outcomes, but legacy estates, data fragmentation, and compliance obligations slow progress.

Blue Origin TeraWave LEO/MEO Network for Enterprises

Blue Origin TeraWave LEO/MEO Network for Enterprises
TeraWave combines 5,280 low Earth orbit satellites with 128 medium Earth orbit satellites—5,408 spacecraft in total—tied together via optical inter-satellite links. The design targets global coverage with two distinct performance tiers: up to 144 Gbps symmetrical RF links per enterprise customer using Q/V-band in LEO, and optical links in MEO delivering up to 6 Tbps for high-throughput trunking between hubs. Blue Origin positions the service for point-to-point private links and enterprise-grade internet access, with an initial target of up to 100,000 customers. The company intends to launch on its own New Glenn vehicles and leverage reusable engines to scale deployment.

NVIDIA Nemotron 3: Hybrid MoE + Mamba‑Transformer for Agentic AI

NVIDIA Nemotron 3: Hybrid MoE + Mamba‑Transformer for Agentic AI
As enterprises move from single-model chatbots to collaborative multi-agent systems, the economic and operational burden of reasoning at scale is becoming the dominant constraint. NVIDIA’s Nemotron 3 family introduces open models and tools designed to keep multi-agent systems fast, affordable and inspectable. The models use a hybrid latent mixture‑of‑experts design to activate only a fraction of parameters per token, combining it with a Mamba‑Transformer approach optimized for long sequences. Nemotron 3 Nano is a small, roughly 30B‑parameter model that activates up to 3B parameters per token, making it efficient for retrieval, summarization, assistants and software debugging.

IBM to Acquire Confluent for Real-Time Enterprise AI

IBM to Acquire Confluent for Real-Time Enterprise AI
IBM has agreed to acquire Confluent for $31 per share in cash, signaling a decisive move to make real-time, governed data the backbone of generative and agentic AI across hybrid cloud environments. The transaction values Confluent at an enterprise value of roughly $11 billion, with closing targeted by mid-2026 pending shareholder and regulatory approvals. Together they aim to unify application, data, and AI pipelines across public clouds, private data centers, and edge locations—reducing integration friction and accelerating time to value for enterprise AI.

Federal AI Order to Preempt State Laws

Federal AI Order to Preempt State Laws
The administration plans an executive order to set a single national AI rulebook and override state-level frameworks, a move with immediate implications for telecom, cloud, and enterprise AI strategies. President Trump signaled he will sign an executive order establishing a uniform federal approach to AI governance that preempts state regulations. Reports indicate the order aims to reduce compliance friction by replacing diverse state rules with a lighter-touch national framework focused on competitiveness. State officials from both parties, safety advocates, and labor groups are preparing to fight the order, citing risks related to consumer harm, deepfakes, hiring bias, and child safety. On the other side, Silicon Valley leaders warn that 50-state compliance regimes could deter innovation and blunt national competitiveness.

AWS invests $50B in federal AI infrastructure

AWS invests $50B in federal AI infrastructure
Amazon Web Services plans a sweeping expansion of classified and government cloud capacity to accelerate AI and high‑performance computing for U.S. agencies. AWS will invest up to $50 billion starting in 2026 to deliver purpose‑built AI and HPC infrastructure for federal customers. The buildout spans AWS Top Secret, AWS Secret, and AWS GovCloud (US) Regions. The expansion is designed to compress analysis timelines and enable AI‑assisted workflows across national security and civil missions. AWS is making a generational bet that AI and HPC, delivered inside accredited government regions at massive scale, will redefine how federal missions operate.

India 6 GHz Spectrum: Wi‑Fi vs IMT Showdown

SPECTRUM CONTROVERSY THE ROLE OF THE 6 GHZ BAND IN INDIA’S TELECOMMUNICATIONS FUTURE
A high-stakes policy fight has emerged in India over the 6 GHz band, pitting global device and cloud ecosystems against mobile operators over whether the band should power unlicensed Wi‑Fi or licensed mobile (IMT) networks. Apple, Amazon, Cisco, Meta, HP, and Intel have jointly urged India’s regulator, TRAI, to reserve the full 6 GHz range for Wi‑Fi, arguing the band is not technically or commercially ready for IMT and that unlicensed use will deliver immediate, widespread capacity benefits. Reliance Jio, Bharti Airtel, and Vodafone Idea have countered that delicensing upper 6 GHz would permanently foreclose India’s option to deploy wide‑area licensed broadband in prime mid‑band spectrum.

WISP vs Starlink Speeds: Ookla 2025

WISP vs Starlink Speeds: Ookla 2025
New performance data shows U.S. WISPs are getting faster, but low‑Earth orbit players like Starlink are advancing just as quickly—and the competitive gap in rural markets is narrowing. Based on Speedtest Intelligence data from Q1 2021 to Q2 2025, eight of the larger WISPs—Starry, Resound Networks, Nextlink, Wisper Internet, Unwired Broadband, GeoLinks, Etheric Networks, and Rise Broadband—improved speeds, with download gains outpacing uploads. Starry led by a wide margin with a 202 Mbps median download in Q2 2025, followed by Resound at 99 Mbps and Nextlink at 68 Mbps; GeoLinks trailed at 23 Mbps. Crucially, only a minority of WISP users consistently achieve the FCC’s 100/20 Mbps fixed broadband benchmark.

Partner Hubs

Download content, access intelligence tools, and hear from executives.

Partner Events

  • FutureNet Asia 2026
  • Network X Vienna 2026
Scroll to Top