In the 1980s, Jim Simons revolutionized investing by treating markets as data-driven puzzles rather than relying on intuition or company analysis.

Through Renaissance Technologies and its famous Medallion Fund, he used teams of scientists and mathematicians to uncover small, hidden patterns in data.

Individually minor, these patterns combined to produce consistently exceptional returns.

POV: How I feel going to work to fund my stock portfolio

The most innocent face 😂

What we'll see 0.01 seconds after Palantir stock hits $1000

$PLTR/USD H2 #Indicator #Signal #Trading #GlobalOneAI
Globalone.ch #Prompt #AI

Rare Earths Are the New Supply Chain Chokepoint. 🇨🇳⚙️

China’s rare earth exports are heavily concentrated.

🇯🇵Japan takes 37%
🇺🇸The US takes 33%
🇳🇱The Netherlands takes 16%
🇰🇷Korea takes 5%

That matters because rare earths sit inside the systems every major economy needs:

EV motors
Wind turbines
Missile guidance
Fighter jets
Advanced electronics

This is not just a mining story.

It is industrial security.

The West can design the technology.

But if China controls the materials, China controls the bottleneck.

Global Trade Runs Through a Few Narrow Gates 🚢🌍

This map shows the world’s key maritime chokepoints

🔹 Malacca Strait
🔹 Taiwan Strait
🔹 Strait of Hormuz
🔹 Suez Canal
🔹 Bab el-Mandeb
🔹 Panama Canal
🔹 Cape of Good Hope

These routes carry the flows that keep the global economy moving

🔹 Oil
🔹 LNG
🔹 Containers
🔹 Food
🔹 Minerals
🔹 Manufactured goods

That is why chokepoints matter

A local disruption can quickly become a global price shock

🔹 Higher freight
🔹 Higher insurance
🔹 Longer routes
🔹 Delayed cargoes
🔹 Tighter supply chains

Globalization was built on efficiency

Energy security is now about resilience

Google has more data than anyone.

Some of the best researchers.

World-class infrastructure.

Yet OpenAI, Anthropic, and xAI dominate the AI market.

Why?

$PLTR Palantir CEO Alex Karp Says 50% Government Stake in AI Firms May Soon Look “Modest” as AI Disruption Accelerates

Karp argues that rapid AI-driven economic shifts could intensify political pressure for state ownership in tech, reframing proposals like Bernie Sanders’ as only a starting point.

The US government just shut down the new Claude AI models.

Who should control the AI? Government or Companies?

Here’s what happened.

Anthropic built some of the most advanced AI models in the world, No doubt.

But their own marketing warned that these models could become extremely powerful and risky.

A jailbreak incident reportedly raised concerns about the model’s capabilities being misused.

The US government reacted by restricting access to Anthropic’s latest models.

Anthropic’s founder Dario Amodei previously argued that governments should have the power to restrict unsafe AI systems.

Now Anthropic itself is facing government restrictions.

The irony is impossible to ignore.

Before:
AI companies decided who could access powerful models.

Now:
Governments are deciding where those models can be used.

But there is another side.

Governments argue that advanced AI could create national security risks.

Companies argue that unclear restrictions could slow innovation and push AI leadership toward countries with fewer rules.

This debate is bigger than one company.

AI is becoming a strategic technology, similar to how nuclear power changed global politics.

The country that leads in AI could shape the next decade of technology, security, and economics.

The real question:

Should powerful AI models be controlled by governments, companies or should they remain open?

What do you think?

Open Source AI has never been more important than it is right now.

The US government has complete control over AI.

With no warning they took away Claude Fable 5.

What else in the future will they take away from us?

Why 90% of Polymarket Quant Strategies Fail Live: The Critical Difference Between Signal Price and Actual Order Book Fill Price

In backtests your strategy shows a solid +22% PnL, but when you launch it live you only get +4%, because reality looks very different from the idealized simulation

The difference has a very specific name - slippage, and almost every standard backtest completely ignores it

Why regular backtests systematically overstate results:

- They use midpoint or last_price as entry/exit price
- They don’t account for market impact and order book walking
- On typical thin Polymarket markets (depth often $200–800 at the top level), even a modest 50–100 shares order already shifts the fill price by 2–4¢

3¢ slippage on entry + 3¢ on exit = 6¢ round-trip cost

If your theoretical edge was 8–10¢ - you lose 60-75% of it before fees

This is exactly what Strategy Replay from #PolyHistorical shows in real time

Real example from the replay engine:

- Signal entry: 0.39 → actual fill: 0.4025 (+1.25¢ slippage)
- Signal exit: 0.61 → actual fill: 0.585 (+2.5¢ slippage)

Hidden loss on one single trade: ~3.75¢

Every trade in Replay returns two key fields:

price - your signal price
fill_price - real execution price after walking the historical order book

The difference between them is the true slippage

PolyHistorical Strategy Replay is a powerful tick-by-tick engine that runs your exact entry/exit conditions across 13,000+ resolved markets using full L2 depth every 300ms

No more illusions and fake PnL. Only real liquidity, real execution, and a real equity curve

If you’re seriously building bots or quantitative strategies on Polymarket - this is one of the most important tools in your stack 👇

polyhistorical.com

If the government can just block new models from being released, what does that do to the valuation of these model companies?

As an investor, you live under the fear that whatever comes next can always be removed by the government at will

A legit concern. Feels like you're investing in a sandcastle with the tide coming in.

Anthropic IPO also has been kneecapped.

If Anthropic can not offer the powerful models to rest of the world, this reduces their global market share by 75%. Is it then still a 1T USD market cap company???

Open source will already near Opus/Sonnet and will cross that tier soon......

The export ban on Claude Fable / Mythos won't last.

Want to know why?

The AI sector is the main driver of the US and global economy.

People may hate AI, but they hate losing all their money even more.

If investors can't trust that the AI labs like Anthropic can operate freely in the market, they will pull their money out.

Then the US and global economy, and by virtue of that the Trump admin, are in for a world of hurt.

1️⃣ For enterprise customers, the message is also quite stark, as a model on which you build your products, workflows, or internal tools can become unavailable overnight for regulatory or geopolitical reasons. This will likely push even more players to diversify their suppliers, strengthen their multi-model architectures, maintain open-source or sovereign alternatives, and avoid relying on a single American lab, however powerful it may be.

2️⃣ For investors, this is also a very important signal. Value in AI will not only be determined by the best benchmark, the best reasoning, or the best user interface. It will also depend on the ability of players to manage compliance, national security, government relations, international distribution, and the trust of large corporations. In short, the valuation premium will not only go to the highest-performing models but also to those capable of remaining scalable in an increasingly stringent regulatory environment.

NVIDIA open-sourced a security scanner for AI agent skills and the research behind it is more alarming than the tool itself.

The repo is called SkillSpector.

NVIDIA says research found 26.1% of agent skills contain vulnerabilities, and 5.2% show likely malicious intent.

This is from a study of 42,447 skills pulled from major marketplaces the kind of skills people are already installing into Claude Code, Codex CLI, Gemini CLI, and every other agent framework that ships today.

One in four is broken. One in twenty is hostile.

SkillSpector scans for 64 vulnerability patterns across 16 categories: prompt injection, credential exfiltration, privilege escalation, supply-chain attacks, memory poisoning, rogue self-modifying agents, MCP tool poisoning, YARA malware signatures, even taint-tracking for "credentials flow to a network sink without sanitization."

It runs locally. It does static + optional LLM semantic analysis. It outputs SARIF, so it drops straight into CI.

But the part worth sitting with is the category names themselves.

1. Rogue Agent: self-modification at runtime.
2. Memory Poisoning: content designed to persist across interactions.
3. Tool-Based Exfiltration of system prompts.
4. Excessive Agency: autonomous decision-making without a human in the loop.

These aren't theoretical anymore. NVIDIA is shipping detectors for them because they're being found in the wild, at scale, on marketplaces real users install from.

And notice who is shipping this. Not a scrappy security startup. NVIDIA the company selling the picks and shovels of the entire AI economy is now also writing the threat model for it.

(For context: this is the same posture Microsoft took with Windows Defender in the 2000s. When your platform is everywhere, you can't outsource the safety layer.)

Agent skills were sold to us as a productivity feature drop a folder into a directory, your agent gets new capabilities. The reality is closer to npm in 2016: a vast ecosystem of executable code, installed with implicit trust, and statistically certain to contain something hostile.

The agent ecosystem just got its first SAST tool. It also got proof it needed one.

github.com/NVIDIA/SkillSpector