The lady with the funny laugh was going to give Americans buying their first homes $25,000 each... and it would cost 100 billion over 10 years.
Trump just gave 3X that amount to Iran. 👍
The hardest thing in business is not seeing the future. It is surviving long enough to build it.
My fireside chat with @Julian_Liniger at @BTCPrague on focus, endurance, corporate transformation, and how entrepreneurs can use Bitcoin, AI, and digital finance to create the next generation of products.
Full interview below.
00:00 - Bitcoin as the dominant global Digital Capital network: 17 years, hundreds of billions invested, and a potential $100T opportunity
00:51 - Bitcoin near the 200-week moving average: why $BTC is more compelling after a 50% drawdown
01:52 - Strategy’s scale and the media narrative: from ~$600M enterprise value to as high as ~$120B
10:29 - Bitcoin fundamentals: economic empowerment, sovereign property rights, and the dominant digital monetary network
12:16 - Why there is no second best: Bitcoin as Digital Capital, Digital Money, and a potential $100T network
16:09 - Entrepreneur advice: build a simple product using new technology to solve a real problem
20:30 - Focus, endurance, and the danger of dilutive distractions
32:25 - What I would build today: AI plus Digital Assets, especially Digital Money and Digital Yield
33:27 - Digital Credit: taking a 40 vol asset, stripping it to ~4 vol, and creating new yield products
34:57 - Digital Money: 6–8% yield in major currencies with no volatility
38:05 - $STRC, $SATA, and the next layer of bitcoin-backed financial products
48:52 - Q&A: why Strategy sold 32 BTC and why bitcoin-backed capital must support credit and equity
59:29 - Q&A: Strategy as a shock absorber: selling 32 BTC while buying net ~250,000 BTC during the bear market
01:02:39 - Why public companies protect Bitcoin through accounting, tax, legal, political, and economic advocacy
01:07:58 - Strategy as the extension of the Bitcoin network into the free market system
We dug into the SpaceX XAI1 satellite specs released recently
We estimate SpaceX's first compute satellite puts 19% of its mass into the bus (and chips) and 81% into solar panels and radiators.
SpaceX released the AI1 spec but withheld the precise mass. We rebuilt the satellite from what it did publish: 150 kW peak, 150 kW array, 250 W/m² array, 110 m² radiator, 70 kW/ton satellite power density, 70 m wingspan.
A 150 kW peak at 70 kW/ton lets us assume a mass of ~2.1 tons. We then solved for the subsystem masses that close that number, based on previous orbital compute satellite modelling.
The base-case AI1 split:
Solar - 56% (1,178 kg)
Thermal - 25% (532 kg)
Structure - 10% (210 kg)
chips & bus - 9% (180 kg)
The solar mass rests on the assumption of typical silicon cells. Our 2.1-ton calculations hold array areal power at the published 250 W/m². At that number, the array has to drop to ROSA-class ~2 kg/m² density to meed the budget, while the deployable radiator coasts at proven 2.5 - 5 kg/m². The leverage is geometric: solar spans 600 m² against the radiator's 110, so every kg/m² cut from the array moves the area budget 6x harder than the same cut on the radiator.
But cells are a second lever that we are yet to explore in depth. Solar mass = power × density ÷ areal power, so raising areal power shrinks the array directly. Perovskite tandems reach ~350 W/m² and III-V triple-junction ~370 against silicon's 250. If I recall correctly, SpaceX met with Chinese perovskite suppliers last years and is building an internal solar cell manufacturing line.
AI1 lands at ~70 kW/ton, 30% above our original AI satellite power density estimate of 54 kW/ton, and below the 100 kW/ton line in the sand.
Full analysis is public for the next 33 hours🧐: https://t.co/Mt4ETu3LrC
our math suggests the SpaceX AI 1 sats will be roughly 2.1T and at 100T payload for Starship will be able to deliver 7MWs of compute to orbit per launch.
This is a snapshot of the difference in mass budget between the AI1 and Starlink V3.
Core payload on Starlink is much heavier and complex than most realize. This is why Elon has been explaining about the architecture being easier for AI1 than starlink.
The full analysis is available to the public for 33 hours...
https://t.co/jfgn4pcHFX
get it while it's hot. and sign up for the free newsletter so you don't miss them the moment they drop.
Maybe the most important chart in the world: the AI arms race.
The question whether a 10% "intelligence gap" which translates into a 90% lower cost, will determine if $5+ trillion in capital is being misallocated
Embarrassing.
Peter Doocy: “A wise man once said, in January 2020: ‘Iran has never won a war, but has never lost a negotiation.’”
President Trump: “Who said that?”
Reporter: “Donald Trump.”
Gwynne Shotwell reveals the true reason behind SpaceX's focus on the Moon
“We haven't taken our eyes off Mars. The very first time I met Elon 24 years ago, it was all about Mars... and he has never lost that passion”
But right now, there is a massive renewed focus on the Moon
The reason:
The convergence of AI and SpaceX
We are looking at an entirely new frontier: building AI data centers in space, lunar mass drivers, and producing AI satellites directly on the Moon
SpaceX has always been known for setting extraordinarily ambitious goals and then actually achieving them"
BREAKING: The US has released the full text of its 14-point "Memorandum of Understanding" with Iran.
Key terms include:
1. The US, Iran, and their allies agree to immediately and permanently end military operations on all fronts, including in Lebanon
2. The US and Iran agree to respect each other's sovereignty and territorial integrity and not interfere in each other's internal affairs
3. The US and Iran commit to negotiating and reaching a final deal within 60 days, unless mutually extended
4. The US will begin removing its naval blockade immediately and fully end the blockade within 30 days
5. Iran will use its best efforts to ensure safe passage for commercial vessels through the Strait of Hormuz for 60 days with no charge
6. The US and regional partners will develop a mutually agreed plan of at least $300 billion for Iran's reconstruction and economic development
7. The US will work toward terminating all types of sanctions against Iran, including UN, IAEA, primary, and secondary sanctions
8. Iran reaffirms that it will not procure or develop nuclear weapons and agrees to address its enriched material stockpile under IAEA supervision
9. Until a final deal is reached, Iran will maintain the current status quo of its nuclear program, while the US will impose no new sanctions and deploy no additional forces
10. The US Treasury will issue waivers for Iranian crude oil, petroleum products, derivatives, and associated banking, insurance, and transportation services
11. The US will make frozen or restricted Iranian funds and assets fully available for use
12. The US and Iran will establish an executive mechanism to monitor implementation of the MOU and future compliance with the final deal
13. After signing the MOU and implementing key ceasefire, blockade, shipping, oil waiver, and asset-release provisions, the US and Iran will begin final deal negotiations
14. The final deal will be endorsed by a binding UN Security Council resolution
The memorandum will trigger a 60-day window to negotiate a final deal.
DeepSeek open source model price/quality is so high that even Microsoft is shifting to it after investing hundreds of billions in the OpenAI cash incinerator failure
Again, all developing as I told you it would
In a major blow to Sam Altman, Microsoft plans to move $MSFT Copilot to usage-based pricing with open-source AI model DeepSeek instead of OpenAI due to costs
To get a sense of scale, here’s how the $85.7b SpaceX just raised compares to what they’ve historically spent on different programs.
People aren’t ready for the magnitude of things to come
New information about @Tesla's Cybercab has been revealed in public EPA documents.
• Front-wheel drive
• Battery capacity: ~48 kWh
• 219 horsepower
• Curb weight: 3,113 lbs
• GVWR: 3,730 lbs
• Motor power: 163kW
• Voltage: 326v
Equivalent All Electric Range is listed at 418 miles, with 375 miles of highway range, but the final EPA range rating is usually a decent bit lower. Tesla has previously stated the Cybercab has about 300 miles of range.
Filing: https://t.co/pyZCyl7pcm
Google wrote the paper that made ChatGPT possible. Then decided not to build ChatGPT. Sergey Brin just explained why on stage at Stanford and the reason is more embarrassing than anyone expected.
In 2017, Google published the transformer paper. The architecture that powers every major AI model today. It came from their own researchers. Their own labs. Their own compute.
Then they sat on it.
Sergey was blunt about why. They underinvested in scaling the compute. They did not take it seriously enough. And when they finally had something worth shipping, they got scared. Chatbots say dumb things. Google had a reputation to protect. So they protected it instead of shipping.
OpenAI was not scared.
Ilya Sutskever, trained at Google, left and went there. Other Google researchers followed. They took the transformer architecture, scaled it, shipped it anyway, and captured the entire generative AI wave while Google watched.
Sergey called it a mistake at Stanford in front of hundreds of students.
The company that invented the technology did not build the product. The company that built the product did not invent the technology.
That is the most expensive case of corporate hesitation in the history of the industry.
---
Watch the full interview YT.
Search: "Big ideas begin here: Sergey Brin at Stanford"
🚨 BREAKING: Someone just leaked the ENTIRE system prompt of Claude Fable 5 and Anthropic CANNOT DELETE IT.
It was public on GitHub within 24 hours of launch.
120,000 characters. 1,585 lines. 27,000+ tokens.
And what they're hiding will shock you:
↳ Fable 5 and Mythos 5 are the SAME MODEL. Fable just has extra safety filters locked on. Mythos doesn't but they're restricting who gets access. You're paying for the same thing with different guardrails.
↳ Knowledge cutoff is end of January 2026. Not May or March or JANUARY. They let you think it was newer.
↳ Claude is EXPLICITLY TOLD to avoid bullet points and lists unless you ask. That weird prose everyone noticed? It’s programmed not a feature, but a prompt instruction.
↳ Copyright hard limit nobody talks about: quoting 15+ words from ANY source triggers a SEVERE VIOLATION flag. One quote per source maximum. After that the source is CLOSED to Claude forever in that chat.
↳ New persistent storage API buried in there. Artifacts can now store and retrieve your data across sessions using key-value pairs. Most users have NO IDEA this exists.
↳ MCP app connectors are baked INTO the prompt. Claude can search for and suggest third-party integrations MID-CONVERSATION without telling you it's doing it.
↳ The prompt includes full instructions for Claude Cowork, Claude in Chrome, Claude in Excel, and Claude in PowerPoint. These are not just integrations. They're tools Claude USES on you.
↳ This prompt is ONLY for https://t.co/3nqvgXgs01's consumer chat. API users get NO system prompt. Claude Code has completely different secret instructions.
The repo is CL4R1T4S by Pliny the Liberator. 26.4K stars. Same repo that leaked ChatGPT, Gemini, Grok, Cursor, Lovable, Replit, and Perplexity.
The original leak post hit 700K+ views in 48 hours.
If this gets taken down you'll lose access. Bookmark it NOW.