Meta has signed a new deal to spend an additional $21 billion with CoreWeave between 2027 and 2032, on top of a prior $14.2 billion commitment. Total exposure: $35.2 billion to a single GPU infrastructure provider. Meta’s projected capital expenditures for 2026 alone sit at $115 to $135 billion — nearly double 2025 levels.

The logic is portfolio hedging. Meta is building its own facilities (including a major Texas data center) while simultaneously contracting with CoreWeave for scalable capacity it can access immediately without the 18-month construction timeline. CoreWeave benefits by diversifying away from Microsoft, which previously represented a dominant share of its revenue.

Microsoft announced a $10 billion investment in Japan spanning 2026 through 2029 — AI infrastructure, cybersecurity partnerships, and a commitment to train over one million engineers and workers by 2030.

It’s the follow-on to a $2.9 billion investment in April 2024. The new package is organized around three pillars: Technology, Trust, and Talent. Microsoft is also joining Japan’s Kyushu Semiconductor Human Resource Development Consortium — the first major international tech company to do so.

Republican Clay Fuller won the special election in Georgia’s 14th Congressional District, replacing Marjorie Taylor Greene, who resigned after breaking with Trump over Gaza, healthcare costs, and the Epstein files.

Fuller is a lieutenant colonel in the Georgia Air National Guard and a former White House fellow during Trump’s first term. He beat Democrat Shawn Harris.

The margin: under 12 points. In a district Trump carried by 37 points in 2024.

North Korea confirmed this week that its latest testing spree included ballistic missiles equipped with cluster-bomb warheads.

That’s a notable escalation in the weapons taxonomy. Cluster munitions scatter submunitions over wide areas — they’re banned under international treaty by over 100 countries (the US and DPRK are not among them). Mounting them on ballistic missiles combines range and penetration with area-denial capability.

Pyongyang framed the tests as part of its ongoing push to expand nuclear-capable forces. Multiple new systems were reportedly involved across the week’s testing schedule.

North Korea confirmed this week that its recent testing spree included ballistic missiles armed with cluster-bomb warheads. Pyongyang described the launches as part of an ongoing push to expand its nuclear-capable forces and diversify its weapons systems.

Cluster munitions on ballistic missiles is a meaningful escalation in the capability profile, not just the quantity. A cluster-bomb warhead transforms a precision ballistic strike into an area-denial weapon — useful for hitting military formations, airfields, or ports rather than single hardened targets. It fills a gap in North Korea’s conventional order of battle while remaining deliverable on platforms already developed for nuclear payloads.

OpenAI has crossed $25 billion in annualized revenue and is reportedly taking early steps toward a public listing — potentially before the end of 2026.

Anthropic is not far behind, approaching $19 billion in annualized revenue. Both numbers represent growth that would have seemed implausible two years ago.

For context: OpenAI’s revenue in 2023 was around $1.6 billion. The AI model market has compressed a decade of typical SaaS scaling into about 24 months. That’s not organic enterprise adoption — it’s a platform shift, and the numbers are reflecting that.

OpenAI has passed $25 billion in annualized revenue and is reportedly taking early steps toward a public listing, potentially as soon as late 2026. The company is currently valued at $852 billion — not a typo. Rival Anthropic is approaching $19 billion in annualized revenue.

An OpenAI IPO would be unlike anything the public markets have processed in years. The valuation, the opacity of the business model, the nonprofit-to-capped-profit structure, and the political entanglement — Greg Brockman and other executives have donated heavily to Trump-aligned super PACs — all make this a strange animal for traditional equity analysis.

One of quantum computing’s central problems is that errors happen and you often don’t know it until you measure — by which point the quantum state is gone. A new method published this week can detect quantum information loss more than 100 times faster than previous approaches, tracking changes in near real time.

That’s not a fix for decoherence. But it’s a significantly better diagnostic tool, which matters for the engineering cycle. You can’t fix what you can’t see.

Samsung reported that first-quarter profit likely surged more than 700% year-over-year, driven by explosive demand for high-bandwidth memory chips used in AI training and inference systems. The number is staggering but follows a period when Samsung’s memory business was being undercut by inventory gluts and pricing pressure.

The reversal is a clean signal: the AI buildout has moved deep enough into the stack that it is now pulling on memory, not just logic chips. Nvidia gets most of the narrative oxygen when people talk about AI hardware, but the demand chain runs straight through DRAM and HBM suppliers. Samsung, SK Hynix, and Micron are all beneficiaries of the same underlying dynamic.

Two weeks in and the Iran ceasefire is fraying at every seam. Iranian parliament speaker Mohammad Bagher Ghalibaf listed three violations he says have already occurred: continued Israeli strikes in Lebanon, a drone entering Iranian airspace, and what Tehran frames as a denial of its right to enrich uranium. None of these are minor complaints — each one is a potential pretext for Tehran to walk away.

The White House’s position isn’t helping. JD Vance, speaking from Hungary, acknowledged the ceasefire is “messy” but held firm that Iran cannot enrich uranium and that Lebanon was never included in the agreement. Iran says the opposite. That is not a gap — that is a chasm.