🇪🇺🇫🇷 France is Europe’s sovereignty model

No other EU country comes close.

🔸 France has around 290 nuclear warheads and no permanent US military bases.

🔸 More than 95% of its electricity came from nuclear and renewables.

🔸 Non-European investment faces review at 10% in listed strategic companies.

This is the model Europe must apply at the continental level

Sovereignty requires energy independence, domestic defence production, business competitiveness and control over strategic sectors.

‼️🇷🇺 Putin announces structural changes within Russia's Ministry of Defense

▫️All logistics services will now be under single command — Valery Solodchuk has been appointed to the position

▫️New military branch for drone warfare will be launched — Denis Lyamin has been appointed to the position

▫️Center Group of Forces — Andrei Ivanayev has been appointed to the position

▫️Vostok Group of Forces — Pyotr Bolgare has been appointed to the position

Chinese Yuan hits strongest level against the U.S. Dollar since February 2023 📈 📈

Game theory explains a little-known paradox: we never lie as much as when we lack options.

Lying is the reflexive strategy of someone who is betting their position on a single move. Without alternatives, manipulation becomes a tactical necessity to compensate for a weakness.

They lie out of fear of losing the only card they hold. Someone who has options, on the other hand, can afford the luxury of truth. Losing one card doesn't destroy them; they have others.

Honesty isn't always a moral quality; it's often the privilege of those who have nothing to lose by being transparent. Don't measure a person's worth by their loyalty, but by the number of options they have when they choose to be loyal.

Nasdaq futures are reversing right at the end of the diamond pattern

The only thing saving it right now is the 60 day moving average.

Russell 2000 stocks above the 200 day moving average finished today back on the ascending trendline.

It's being crushed in the corner of the wedge that ends next week.

Is this warning us a commodity and market selloff is coming?
Or is it about to break the 2018 resistance line?

GPT-5.6 Luna Max is the best value model on the market right now.

One of the best things you can do is use GPT-5.6 Sol as the orchestrator and spin up subagents with Luna Max.

This prompt does exactly that and cuts token costs by 50% or higher

Traders only have 4 moods

If you're capable of playing the long game, you already possess an asymmetrical advantage over your opponents. The majority of people suffer from a chronic condition: addiction to immediate gain. They sacrifice their future for instant gratification. By sitting back and building your foundations while they're busy making quick fixes, you're simply waiting for them to burn themselves out.

There exists a category of individuals devoid of any particular talent, without genius or superior IQ, who nevertheless rise to the top of the hierarchy.

Their secret? An implacable will to dominate any situation, any adversary, to obtain what they seek, without ever giving up.

It was this predatory temperament that characterized the early career of John Rockefeller.

Trying to win a game with no finish line is a fast track to resource exhaustion.

#simonsinek dropped one of the most practical game theory breakdowns ever:

Finite players play to beat a rival, hit a quarterly metric, or win a war. infinite players play for one reason only: to stay in the game forever.

Key takeaways in the article.

This article teaches you everything you need to know about graph engineering

MAJOR CORRECTION IN MEMORY STOCKS AFTER SANDISK AND WESTERN DIGITAL EARNINGS

Entire sector is crashing in the premarket:

- $WDC down 16%
- $SNDK down 12%
- $STX down 6%
- $SKHY down 7%
- $MU down 6%

Both companies beat revenues. SanDisk revenue was up 372% year over year, Western Digital up 44%.

The problem was guidance. SanDisk grew 51% last quarter and guided to 18% for the next one.

Growth is slowing, and the market thinks memory prices have peaked.

Too many choices will kill your momentum faster than outright laziness.

Warren buffett told his pilot to list his top 25 goals, circle the top 5, and treat the remaining 20 like a disease to be avoided at all costs.

Execution isn't about deciding what to do. it's about aggressively locking out everything else so only one path remains.

This billionaire hedge fund manager studies 272 charts every day.

Smart people end up making stupid group decisions for one simple reason:

The rules of the game force them to.

Dr ashley hodgson breaks down why team projects fail, why industries ruin their own markets, and why good people turn dirty. when the system rewards the first person who cheats, playing nice isn't honorable, it's suicide.

This is one of the best explanations of the prisoner's dilemma i've come across.

Key concepts in the article

Prime Intellect just open-sourced Prime Agent.

It's a self-improving coding harness that scores 95.5% on ARC-AGI-3 - well above the human baseline.

This new agent is insane and can rewrite its own prompts, memory, and skills mid-task.

Here's how to install it in <60 seconds:

Step 1. Run the install command

curl -fsSL app.primeintellect.ai/prime-agent/install| sh

Works on Linux or macOS.

Step 2. Connect a model

Prime Agent works with whatever you already have access to:

→ Subscription logins: Codex, Claude Pro/Max, GitHub Copilot
→ API keys: Anthropic, OpenAI, Google, Groq, Fireworks, Prime Inference, and more
→ Self-hosted: vLLM, Ollama, LM Studio, or open-weight models like GLM-5.2 if you want everything running inside your own network

Step 3. Start a session

Once connected, just start working. It runs as a persistent Python environment, so it can call tools, launch subagents, and reference its own history without losing context.

Step 4. Let it self-improve as it works

Use the /refine command mid-task, and it rewrites its own prompts, memory, and skills based on what's actually working in that session.

That's it. MIT licensed, fully free, works with open and closed models alike.

→ github.com/PrimeIntellect-ai/prime-agent

Prime Intellect's new AI agent just scored ABOVE human intelligence on ARC AGI.

95.5, which puts it well above the average human.

AGI is here?

primeintellect.ai

U.S Treasury raised its Q3 government borrowing estimate to $739 billion, which will push the national debt over $40 trillion!

This is $68 billion more ​than it projected in May, 2026

In Q4, Treasury is expected to borrow an additional $628 billion, based ​on a year-end cash balance of $850 billion.