🚨 JUST NOW: President Trump signs TWO executive orders securing the United States' spot as the WORLD LEADER in Quantum Computing
ORDER 1: Launches a national effort to produce a quantum computer capable of performing important scientific calculations. It also aims to develop quantum-enabled sensors and networks within the next 5 years.
ORDER 2: Directs federal agencies to transition their computer systems to quantum cryptography (post-quantum cryptography standards) by 2031. It positions the U.S. government to lead wider adoption of these stronger security standards to address future quantum computing threats
La primera mujer de Elon Musk contó una vez cómo es verlo fracasar.
Dijo que no reacciona como una persona normal. Cuando un cohete explota, la mayoría de la sala se queda en silencio. Algunos lloran. Otros empiezan a calcular cuánto dinero acaban de perder.
Musk saca el móvil y empieza a hacer llamadas. No llamadas emocionales. Llamadas de ingeniería. "Qué ha fallado. Cuándo lo arreglamos. Cuándo es el próximo lanzamiento." Su voz no cambia. Su cara no cambia. El cohete que acaba de costar 60 millones ya es pasado. Lo único que existe es el siguiente.
Dijo que fue lo más perturbador que había visto en su vida. No porque fuera frío. Porque de verdad no le afectaba. El fracaso no se registraba como fracaso. Se registraba como datos. Un experimento que produjo resultados. Resultados que informan el siguiente experimento.
Por eso gana. No porque no fracase: fracasa más espectacularmente que nadie en la historia. Gana porque el fracaso ocupa cero espacio psicológico. Entra como dato y sale como acción.
La mayoría no pierde por fracasar. Pierde porque se pasa semanas procesando el fracaso antes de volver a actuar. Musk se pasa cero segundos. La distancia entre el fracaso y el siguiente intento es una llamada de teléfono.
Y esto es lo que casi nadie entiende: esa frialdad no es un don. Es una forma de pensar. Y las formas de pensar se entrenan.
Por eso, el problema no es que fracases. Es cuánto tiempo dejas que el fracaso viva dentro de tu cabeza.
En mis lista privada (https://t.co/U7WqyGNiJb) comparto modelos mentales como el de separar el dato de la emoción, que te pueden ayudar a recuperar más control sobre tus acciones.
I’m 60 years old and retired from JPMorgan. My monthly income is $105,000.
My June advice 22th:
$PLTR (Palantir) — Don’t buy
$SNDK (SanDisk) — Don’t buy
$CRWV (CoreWeave) — Don’t buy
$AXTI (AXT Inc) — Buy at $70–77
$INTC (Intel) — Buy at $122–$128
$SPCX (SpaceX) — Buy at $160–$165
$MU (Micron Technologys) — Buy at $1,060–$1,080
People ask, Why don’t you charge?
I’ve made enough. Sharing is my passion ,that’s why I post for fre.
Elon Musk just signed another blockbuster compute deal and it confirms something that is reshaping the entire AI infrastructure landscape (Save this).
SpaceX signed a deal with open source AI startup Reflection AI, giving them access to Nvidia GB300 chips at the Colossus 2 data center in Memphis.
Reflection will pay $150 million per month starting July 1, totaling up to $6.3 billion if the deal runs through 2029.
Elon has been building what analysts are now calling EWS, Elon Web Services, the fourth major hyperscale AI cloud alongside AWS, Azure, and Google Cloud.
Anthropic committed $1.25 billion per month through 2029 for full access to the Colossus 1 data center over 300 megawatts of compute, 200,000+ GPUs.
Google committed $920 million per month from October 2026 through June 2029 for access to approximately 110,000 Nvidia GPUs.
That is $11 billion per year and up to $30 billion over the life of the contract.
Now Reflection is paying $150 million per month on top of that.
SpaceX now operates as a vertically integrated stack, launch infrastructure, Starlink connectivity, compute, hyperscaler-level cloud services, space data centers, then applications and AI models on top.
That is a civilization scale infrastructure stack that nobody else on earth can replicate.
Elon has three facilities running right now, Colossus, Macro Hard, and Macro Harder.
Colossus 1, built on H100s and optimized for inference, is what went to Anthropic. Macro Hard and Macro Harder are running on Blackwell architecture, 1.2 gigawatts combined and that is where Google is renting capacity as it comes online through the second half of 2026.
He gave the older facility to Anthropic, kept the cutting-edge Blackwell clusters and is now monetizing both.
Analysts had been modeling SpaceX revenue in the mid-$20 billion range.
The compute leasing business alone is now projected to generate approximately $45 billion in incremental annual revenue on top of that.
That revenue directly subsidizes the continued buildout of Grok, future Colossus expansions, and SpaceX's orbital data center ambitions without xAI needing to deliver immediate product revenue to justify the infrastructure spend.
This solves the biggest question mark in the SpaceX IPO story.
The concern going into the roadshow was always, what if Elon builds ahead of demand and xAI can not generate the revenue to justify the investment?
That question is now answered, he does not need xAI to carry the financial load.
The hyperscalers are paying him billions per month to use the compute he built and he is keeping the best hardware for himself.
Milk Road remains bullish on SpaceX/xAI, and we are tracking every deal, every data center announcement, and every compute contract as it happens.
Come join Milk Road Pro and get our full breakdown, the valuation framework, the revenue projections, and exactly how we are thinking about sizing this position.
Link below!