AMERICA'S INTELLIGENCE BRIEFING  ·  UNFILTERED
Patriot Wire  · 

*America's intelligence briefing. Unfiltered.*

Patriot Wire -- April 27, 2026

The Big Picture

Iran is holding a fifth of the world's oil supply hostage through a mined strait, China just rehearsed invading Taiwan with a new class of warship, and Beijing's AI labs are shipping frontier models on Chinese hardware after systematically stealing American research. Three adversaries, three pressure campaigns — all running simultaneously while Washington decides how to respond. The decisions made in the Situation Room this week will echo for years.


Today's Stories

IRAN IS BETTING THE STRAIT AGAINST ITS NUCLEAR PROGRAM — AND WASHINGTON HOLDS THE CARDS

Tehran, through Pakistani mediators, has handed the White House a new proposal: reopen the Strait of Hormuz and end the shooting war, with nuclear negotiations kicked down the road to a "later stage." President Trump is convening his national security team in the Situation Room Monday to weigh the offer. Here's what the proposal actually means: Iran wants to trade away its most powerful economic weapon — a chokepoint carrying roughly a fifth of global oil — in exchange for keeping its nuclear program off the table indefinitely. Brent crude has already surged nearly 17% in a single week, the IEA has called this the largest oil supply shock in recorded history, and American families are paying for every day this blockade holds. Meanwhile, Iran's Foreign Minister Araghchi flew to Moscow after blaming Washington's "excessive demands" for last week's failed talks — a deliberate pivot toward Putin designed to harden Tehran's leverage. The U.S. Navy has turned away approximately 38 ships from Iranian ports. Accept Iran's sequencing and you hand every adversary a playbook: mine a chokepoint, wait out the pain, protect your weapons program. Reject it and the blockade — and $107 oil — stays live.


CHINA REHEARSED THE TAIWAN INVASION. THIS TIME WITH A NEW WARSHIP.

The People's Liberation Army just wrapped "Mission Justice-2025," its latest Taiwan encirclement exercise — and this one was different in a specific, measurable way. For the first time, China deployed a Type 075 amphibious assault ship into a Taiwan Strait drill. That's Beijing's purpose-built marine landing platform — the ship you need not just to fire missiles at Taiwan, but to put boots on the beach. Separately, China's first Type 076 amphibious assault ship, the Sichuan, has entered sea trials in the South China Sea, reportedly testing electromagnetic catapults capable of launching heavier fixed-wing drones. The Pentagon publicly called on Beijing to stop pressuring Taiwan. Words. What matters is that China is no longer just rehearsing the opening missile salvo — it is normalizing the full invasion sequence: blockade, port seizure, beach landing, organic air power. Every exercise that passes without consequence is a green light for the next one. The signal to watch: whether PLA vessels linger near Taiwan's eastern coast after the exercise officially ends.


BEIJING'S AI LABS ARE RUNNING ON CHINESE CHIPS — AND WASHINGTON JUST NOTICED

The White House Office of Science and Technology Policy has formally accused Chinese entities of running "deliberate, industrial-scale campaigns" to steal American frontier AI systems — training Chinese models on the outputs of American ones to skip years of research. The State Department sent cables to embassies worldwide warning foreign governments about alleged IP theft by DeepSeek, Moonshot AI, and MiniMax. Here's the problem: the accusation arrived the same day DeepSeek shipped V4, a model tuned to run on Huawei's Ascend chips — not Nvidia's. The entire U.S. export control strategy was built on the assumption that cutting off advanced semiconductors would slow China's AI development. DeepSeek V4 prices its inference at roughly 90% below American frontier models and runs on Chinese hardware. The strategy has coincided with adaptation, not surrender. As one semiconductor analyst told Reuters: China's top AI models can now run on Chinese chips. We spent years tightening export controls on the wrong layer of the stack.


$4.8 BILLION FOR ROCKETS: THE PENTAGON IS FINALLY BUYING AFFORDABLE MASS

The U.S. Army awarded Lockheed Martin up to $4.79 billion for Guided Multiple Launch Rocket System rounds — the GPS-guided rockets fired from HIMARS that have become the backbone of precision strike from Ukraine to the Pacific. Two full-rate production lots. This contract matters beyond the dollar figure. Western planners have identified an "affordable mass" problem: the Pentagon's habit of buying a handful of perfect, expensive weapons instead of large quantities of proven, cost-effective ones. Ukraine has demonstrated what happens when you run out of the latter. GMLRS is the answer to that problem — reliable, GPS-guided, and suddenly in extraordinary demand. The test now is whether the industrial base can actually deliver. If supplier bottlenecks swallow the funding, "affordable mass" stays a slogan on a PowerPoint slide. Watch for named supplier-expansion announcements in the next 60 days. If they don't come, the munitions shortage that has constrained Western support for Ukraine remains unresolved — and Pacific planners are watching.


YOUR AI TOOLS HAVE A CHINA-SIZED HOLE IN THEM — AND HACKERS FOUND IT IN TEN HOURS

CISA added six new vulnerabilities to its Known Exploited Vulnerabilities catalog this past week. The most alarming: CVE-2026-39987, a flaw in the Marimo Python notebook platform that allows unauthenticated remote code execution — meaning an attacker who finds your server gets a shell with no login required. Per Sysdig's threat research team, the first in-the-wild exploitation attempt arrived 9 hours and 41 minutes after the advisory was published. No public proof-of-concept existed; someone built the exploit during a single workday. From April 11–14, Sysdig observed 662 exploit events across 10 countries, including credential theft, lateral movement, and deployment of malware using blockchain-based command and control. Separately, a threat actor compromised more than 59,000 Next.js servers in 48 hours by chaining two vulnerabilities — a campaign with a reported 64% success rate. American businesses, government contractors, and defense-adjacent firms running AI tooling on the open internet need to understand: the patch window is no longer days. It is hours.


What to Watch


The Closer

Iran is mining a strait, China is rehearsing a beach landing, Beijing's AI labs are running on Chinese chips, and a foreign-built banking trojan just cracked Russia's app store top 200 — all in the same 72-hour window. This is not a series of unrelated events. It is what happens when adversaries read American hesitation as permission. The Situation Room meeting Monday morning may be the most consequential in years. Let's hope the people in that room understand what's actually at stake.


Patriot Wire

America's intelligence briefing. Free, daily, unfiltered.