Daily AI Briefing — July 30, 2026

If you’ve been watching AI news and wondering, “Why are companies spending this much on infrastructure?” — today is a masterclass in the answer. We’re covering seven stories, from a billion-dollar acquisition to a $100 billion campus, a fun Google Earth update, and a provocative vision for the future of AI. Let’s dive in.

1. Nscale to Acquire Anyscale for ~$1.65 Billion

The story: Nscale, an Nvidia-backed cloud provider, signed a deal to acquire Anyscale — the company behind Ray, an open-source framework that helps developers run AI workloads across thousands of chips. Bloomberg reports the price at roughly $1.65 billion. Anyscale’s ~200 employees across the US, Europe, and India will join Nscale. Nscale Press Release | Reuters

Why you should care: Think of Ray as a “traffic controller” for AI computing — it handles the complex choreography of training models across thousands of GPUs at once. By buying Anyscale, Nscale gets the software layer that makes raw compute easy to use. For developers, this could mean tighter integration between cloud hardware and the tools they already work with.

2. Microsoft Azure Crosses $100 Billion; Copilot at 30 Million Seats

The story: Microsoft reported Q4 FY26 earnings with $90 billion in quarterly revenue (+18% YoY). Azure grew 43%, pushing its full-year revenue past $100 billion for the first time. Microsoft 365 Copilot — the AI assistant baked into Word, Excel, and PowerPoint — crossed 30 million paid seats. Microsoft Investor Relations

Why you should care: When a single division hits $100 billion in annual revenue, it’s a milestone. Azure’s 43% growth means companies are migrating to the cloud faster than ever — and AI workloads are a major driver. Copilot’s 30 million paid seats? That’s 30 million people who decided an AI writing assistant is worth paying for. If you wondered whether AI tools are actually being adopted in the real world: yes, and fast.

3. Meta Raises 2026 AI Capex Floor to $130 Billion

The story: Meta reported Q2 revenue of $60.8 billion (+28% YoY) and narrowed its full-year capital expenditure range upward to $130-145 billion, raising the floor from $125 billion. Costs surged 55% to $42 billion, partly from $2.4 billion in legal charges and $1.18 billion in severance from ~8,000 May layoffs. Meta Investor Relations | Yahoo Finance

Why you should care: “Capex” is money spent on long-term assets — data centers, GPUs, networking gear. When Meta says it will spend at least $130 billion this year on these things, it’s telling you that AI isn’t a side project; it’s the core of its future. To put it in perspective: $130 billion is more than the GDP of some small countries. Meta is betting that building massive AI infrastructure today will pay off tomorrow, whether through better recommendation algorithms, new products, or the metaverse.

4. EU Opens Bidding for 7 AI Gigafactories — $11.4 Billion

The story: The European Commission opened a call for proposals to build seven AI gigafactories across Europe, backed by €10 billion ($11.4 billion) in public funding, with a goal of attracting €20 billion more from private investors. Each facility will host at least 100,000 AI chips — roughly four times the scale of current EU data centers. AP News | US News

Why you should care: Right now, most AI computing power is concentrated in the US and China. A gigafactory is a purpose-built facility packed with specialized chips for training advanced AI models. The EU is essentially saying, “We don’t want to be left behind, and we’re spending billions to make sure we aren’t.” For beginners: this is about geopolitical competitiveness as much as it’s about technology.

5. Brookfield and NextEra Plan $100 Billion AI Campus in Kentucky

The story: Brookfield and NextEra Energy announced a $100 billion AI data center campus at the former Paducah Gaseous Diffusion Plant in Kentucky. The plan: 1.2 gigawatts (GW) of computing power by 2032, scaling to 1.8 GW, powered by a paired 2 GW clean energy farm. DataCenterKnowledge | Yahoo Finance

Why you should care: To understand 1.2 GW: that’s roughly the output of a large nuclear plant — enough to power about a million homes. A former uranium enrichment site being transformed into an AI computing campus says everything about how the energy needs of AI are reshaping industrial landscapes. The clean energy pairing is also significant: AI data centers guzzle electricity, and companies feel pressure to make it green.

6. Google Adds Nano Banana Image Generation to Google Earth

The story: Google rolled out a “Create Image” button in Google Earth web, powered by its Nano Banana 2 model. Zoom in anywhere, tap the button, describe what you want to see — and it generates it. Examples include historical recreations (Pompeii in 78 AD), real estate mockups, and custom infographics overlaid on maps. Global rollout started today. Google Blog | Engadget

Why you should care: Not every AI story has to be about billions of dollars. Nano Banana is Google’s lightweight image generation model, and putting it in Google Earth makes exploration more creative and interactive. Want to see your neighborhood with different architecture? Or San Francisco during the Gold Rush? Just type it. It’s a reminder that AI can also make everyday tools more delightful.

7. Zuckerberg Lays Out Meta’s Vision for Personal AI Superintelligence

The story: Mark Zuckerberg published a Wall Street Journal op-ed arguing that AI superintelligence — AI that surpasses human intelligence across every domain — must be made accessible to individuals, not locked inside a handful of institutions. Artificial Intelligence News

Why you should care: “Superintelligence” refers to AI smarter than the smartest human in creativity, reasoning, and everything else. We don’t have it yet, but the debate about who controls it when it arrives is already here. Zuckerberg’s position: it should be democratic, distributed, and personal — everyone gets their own powerful AI assistant, not just big companies. Whether you love or fear that vision, it matters because Meta is spending $130 billion this year to build the infrastructure to make it possible.

The Big Picture

Today’s briefing has one clear theme: the AI infrastructure buildout is accelerating.

• Private industry is spending recklessly. Meta alone could drop $145 billion on capex. Microsoft’s Azure cracked $100 billion. Nscale is paying $1.65 billion for software tools. And that’s just one day.
• Governments are racing to compete. The EU is putting $11.4 billion behind gigafactories. The US is seeing massive projects like the Kentucky campus. No major economy wants to be left without AI computing capacity.
• Consumer AI is quietly maturing. Nano Banana in Google Earth is a small feature, but it shows that AI is becoming a standard part of everyday tools.
• The vision is getting bolder. Zuckerberg’s op-ed argues superintelligence should be personal. That sounds like science fiction, but $130 billion in spending suggests Meta believes it’s just around the corner.

What this means for an AI beginner: The industry you’re learning about is investing like it expects a huge payoff. The infrastructure being built today — data centers, chips, cloud platforms — will power the tools of tomorrow. And the debates happening now about openness, centralization, and access will shape who gets to use those tools and how. Keep learning. The best time to start was yesterday; the second best is today.

See how these tools compare → /comparisons/

📖 Related Reads

• ToolBrain — tool reviews, LLM comparisons, and AI workflow guides

Cross-links automatically generated from None.