This media is not supported in your browser
VIEW IN TELEGRAM
Marc Andreessen is extremely impressed by AI systems.
99% of the time, the answer he gets from the most advanced AI models is better than what he'd get from almost any expert he has access to.
He broke down why using two concepts from psychology:
First: fluid intelligence. The ability to conceptualize problems, process new information, reason through novel situations. The AI has it.
Second: crystallized intelligence. The accumulated knowledge base, the memorized expertise across every field. The AI has been trained on the complete corpus of human knowledge.
AI is a world-class doctor. A world-class lawyer. A world-class accountant, political strategist, marketing expert, and software engineer. All at the same time.
Andreessen's point is that we've crossed a threshold. This isn't about AI being a useful tool anymore. It's about AI being genuinely smarter and better-informed than the best human experts most people have access to.
The disruption this creates in professional services is hard to overstate.
Every industry where people pay for expert judgment is sitting on a fault line.
99% of the time, the answer he gets from the most advanced AI models is better than what he'd get from almost any expert he has access to.
He broke down why using two concepts from psychology:
First: fluid intelligence. The ability to conceptualize problems, process new information, reason through novel situations. The AI has it.
Second: crystallized intelligence. The accumulated knowledge base, the memorized expertise across every field. The AI has been trained on the complete corpus of human knowledge.
AI is a world-class doctor. A world-class lawyer. A world-class accountant, political strategist, marketing expert, and software engineer. All at the same time.
Andreessen's point is that we've crossed a threshold. This isn't about AI being a useful tool anymore. It's about AI being genuinely smarter and better-informed than the best human experts most people have access to.
The disruption this creates in professional services is hard to overstate.
Every industry where people pay for expert judgment is sitting on a fault line.
This media is not supported in your browser
VIEW IN TELEGRAM
Jensen Huang explained why Nvidia's demand is going parabolic.
Two years ago, AI could generate content. Impressive but limited. Then something changed:
Generative AI โ reasoning โ planning โ agentic systems
Agentic AI doesn't just respond to a prompt. It understands a goal, reasons about it, plans steps, uses tools, checks the results, revises the plan and keeps going until the job is done.
Jensen's example: they kicked off a software programming job using an AI agent. It ran for a week. What it produced would have taken a full engineering team a month.
Here's what that does to compute demand.
A regular AI query takes a fraction of a second of computation. An agentic task running for a week takes 100x to 1000x more. Now multiply that by the fact that every enterprise, every company, every developer is deploying agents everywhere.
"Between the amount of computation and the amount of demand use going up, that product is our demand."
The compute requirement isn't growing linearly. It's the product of two exponentials: more computation per task and more tasks being run.
And Jensen's view on where it goes from here:
The great engineer of today works alongside one agent. The great engineer of the future orchestrates hundreds of agents, each orchestrating their own sub-agents.
This is why he says we've arrived at the era of useful AI. The kind that enterprises actually build their operations around.
Two years ago, AI could generate content. Impressive but limited. Then something changed:
Generative AI โ reasoning โ planning โ agentic systems
Agentic AI doesn't just respond to a prompt. It understands a goal, reasons about it, plans steps, uses tools, checks the results, revises the plan and keeps going until the job is done.
Jensen's example: they kicked off a software programming job using an AI agent. It ran for a week. What it produced would have taken a full engineering team a month.
Here's what that does to compute demand.
A regular AI query takes a fraction of a second of computation. An agentic task running for a week takes 100x to 1000x more. Now multiply that by the fact that every enterprise, every company, every developer is deploying agents everywhere.
"Between the amount of computation and the amount of demand use going up, that product is our demand."
The compute requirement isn't growing linearly. It's the product of two exponentials: more computation per task and more tasks being run.
And Jensen's view on where it goes from here:
The great engineer of today works alongside one agent. The great engineer of the future orchestrates hundreds of agents, each orchestrating their own sub-agents.
This is why he says we've arrived at the era of useful AI. The kind that enterprises actually build their operations around.
โ1
How Military & Defense Satellites Are Made in 2026?
1. Mission Design & Requirements
Classified specifications for resilience, anti-jamming, and mission-specific payloads.
Tickers: $LMT ยท $BA ยท $RTX ยท $NOC
2. Radiation-Hardened Components
Building space-grade electronics, processors, and sensors that survive extreme radiation and EMP threats.
Tickers: $RTX ยท $HON ยท $LMT ยท $NOC
3. Bus & Platform Integration
Assembling the satellite bus with secure power, propulsion, and attitude control systems.
Tickers: $LMT ยท $BA ยท $RTX ยท $RDW
4. Payload & Sensor Integration
Mounting advanced EO/IR cameras, SIGINT antennas, laser comms, and AI processing units.
Tickers: $LMT ยท $RTX ยท $NOC ยท $HON
5. Extreme Testing & Qualification
Thermal vacuum, vibration, radiation, and anti-tamper testing in secure facilities.
Tickers: $LMT ยท $BA ยท $RTX ยท $NOC
Military satellite manufacturing in 2026 is classified, ultra-reliable engineering at the highest level โ producing resilient systems that provide strategic superiority in space.
Thatโs why defense giants $LMT, $RTX, $NOC, $BA dominate this high-margin, mission-critical segment.
1. Mission Design & Requirements
Classified specifications for resilience, anti-jamming, and mission-specific payloads.
Tickers: $LMT ยท $BA ยท $RTX ยท $NOC
2. Radiation-Hardened Components
Building space-grade electronics, processors, and sensors that survive extreme radiation and EMP threats.
Tickers: $RTX ยท $HON ยท $LMT ยท $NOC
3. Bus & Platform Integration
Assembling the satellite bus with secure power, propulsion, and attitude control systems.
Tickers: $LMT ยท $BA ยท $RTX ยท $RDW
4. Payload & Sensor Integration
Mounting advanced EO/IR cameras, SIGINT antennas, laser comms, and AI processing units.
Tickers: $LMT ยท $RTX ยท $NOC ยท $HON
5. Extreme Testing & Qualification
Thermal vacuum, vibration, radiation, and anti-tamper testing in secure facilities.
Tickers: $LMT ยท $BA ยท $RTX ยท $NOC
Military satellite manufacturing in 2026 is classified, ultra-reliable engineering at the highest level โ producing resilient systems that provide strategic superiority in space.
Thatโs why defense giants $LMT, $RTX, $NOC, $BA dominate this high-margin, mission-critical segment.
What Kind of Satellites Exist in 2026?
10. Military & Defense: Secure comms, surveillance, high-priority missions โ $NOC, $LMT, $BA
9. GEO Communications: TV broadcast, enterprise backhaul โ $SES, $I, $ETL.PA
8. LEO Mega-Constellations: Global broadband, low latency โ Starlink, Kuiper, $ASTS
7. Earth Observation: Imaging, climate & infrastructure data โ $PL, $MAXR, $BA
6. Weather & Climate: Storm tracking & forecasting โ NOAA, EUMETSAT, JAXA
5. Maritime & Aviation: Tracking ships, aircraft & assets โ $IRDM, $ISAT.L, Thales Alenia Space
4. Agriculture & Resources: Crop monitoring & environmental data โ Sateliot (Private), $BKY, ICEYE (Private)
3. Science & Research: Space telescopes & deep-space missions โ NASA, ESA, JAXA
2. Tech Demo & Education: Small sats for testing new tech โ Tyvak (Private), $SPIR, NASA CubeSat Program
1. CubeSats & NanoSats: Low-cost IoT, research & niche missions โ $PL, Fleet Space (Private), GOMSpace (Private)
Satellites are the hidden infrastructure powering everything from weather forecasts to global broadband.
10. Military & Defense: Secure comms, surveillance, high-priority missions โ $NOC, $LMT, $BA
9. GEO Communications: TV broadcast, enterprise backhaul โ $SES, $I, $ETL.PA
8. LEO Mega-Constellations: Global broadband, low latency โ Starlink, Kuiper, $ASTS
7. Earth Observation: Imaging, climate & infrastructure data โ $PL, $MAXR, $BA
6. Weather & Climate: Storm tracking & forecasting โ NOAA, EUMETSAT, JAXA
5. Maritime & Aviation: Tracking ships, aircraft & assets โ $IRDM, $ISAT.L, Thales Alenia Space
4. Agriculture & Resources: Crop monitoring & environmental data โ Sateliot (Private), $BKY, ICEYE (Private)
3. Science & Research: Space telescopes & deep-space missions โ NASA, ESA, JAXA
2. Tech Demo & Education: Small sats for testing new tech โ Tyvak (Private), $SPIR, NASA CubeSat Program
1. CubeSats & NanoSats: Low-cost IoT, research & niche missions โ $PL, Fleet Space (Private), GOMSpace (Private)
Satellites are the hidden infrastructure powering everything from weather forecasts to global broadband.
๐Stock Stage Analysis: Strongest Themes-Sector-Industry ๐ช
There are 168 stocks with above average ATR that are in the "Strong" Quadrant for all three metrics:
- Strong Sector
- Strong Theme
- Strong Industry
Those stocks represented as stage analysis for swing trading.
There are 168 stocks with above average ATR that are in the "Strong" Quadrant for all three metrics:
- Strong Sector
- Strong Theme
- Strong Industry
Those stocks represented as stage analysis for swing trading.
โ1
The global economy is $126 trillion in 2026.
The US and China alone account for $53 trillion of it 42% of everything. ๐งต
Top economies by 2026 nominal GDP:
๐บ๐ธ United States: $32.4T
๐จ๐ณ China: $20.9T
๐ฉ๐ช Germany: $5.5T
๐ฏ๐ต Japan: $4.4T
๐ฎ๐ณ India: $4.2T
๐ฌ๐ง UK: $4.3T
๐ซ๐ท France: $3.6T
๐ฎ๐น Italy: $2.7T
๐ท๐บ Russia: $2.7T
๐ง๐ท Brazil: $2.6T
The US and China alone account for $53 trillion of it 42% of everything. ๐งต
Top economies by 2026 nominal GDP:
๐บ๐ธ United States: $32.4T
๐จ๐ณ China: $20.9T
๐ฉ๐ช Germany: $5.5T
๐ฏ๐ต Japan: $4.4T
๐ฎ๐ณ India: $4.2T
๐ฌ๐ง UK: $4.3T
๐ซ๐ท France: $3.6T
๐ฎ๐น Italy: $2.7T
๐ท๐บ Russia: $2.7T
๐ง๐ท Brazil: $2.6T
Forwarded from ๐ขWTI, Brent Crude Oil, NGAS Trading ๐
Iran's "Hormuz weapon" is real.
But every year it gets slightly less powerful and a deal may now be closer than at any point since the war began.
But Trump says a deal is "largely negotiated."
Iranian officials reference a 14-point framework.
Structure:
end hostilities โ gradual Hormuz reopening โ 30-60 day nuclear negotiation window.
But every year it gets slightly less powerful and a deal may now be closer than at any point since the war began.
But Trump says a deal is "largely negotiated."
Iranian officials reference a 14-point framework.
Structure:
end hostilities โ gradual Hormuz reopening โ 30-60 day nuclear negotiation window.
๐จ๐ญ Global One SI - Prompt
10 WAYS TO PLAY ROBOTICS IN 2026 1. $TSLA building one of the most vertically integrated humanoid stacks combining Optimus Gen 3 hardware, its in-house AI5 inference chip & same FSD-based AI brain powering its vehicle fleet. 2. $NVDA platform layer for entireโฆ
$OUST
Ouster is the best way to play robotics in 2026 imo.
Beyond its native colour LiDAR and big customers like Amazon, Google etc., the recent acquisition of StereoLabs was a really smart move.
By integrating ZED stereo cameras, Ouster is now also perfectly positioned to capture the growing humanoid robotics market. Very bullish.
Ouster is the best way to play robotics in 2026 imo.
Beyond its native colour LiDAR and big customers like Amazon, Google etc., the recent acquisition of StereoLabs was a really smart move.
By integrating ZED stereo cameras, Ouster is now also perfectly positioned to capture the growing humanoid robotics market. Very bullish.
๐จ๐ญ Global One SI - Prompt
INSANE: A Trump-linked whale opened a $16.8 million oil short hours before a US-Iran peace deal is set to be announced. Peace = lower oil. The trade is perfectly positioned for an event that hasn't happened publicly yet. The size is institutional. Theโฆ
What are we gonna call this? 'Gulf of America' is already takenโฆ
๐จ๐ญ Global One SI - Prompt
What are we gonna call this? 'Gulf of America' is already takenโฆ
Strait of America of course
๐ฟ1
The White House is reportedly nearing a deal with Anthropic to let the NSA and other US intelligence agencies use its advanced AI models for classified work, according to The New York Times.
The talks come despite the Pentagon previously labeling Anthropic a "supply chain risk" after disputes over contract language tied to surveillance and autonomous weapons use.
The report also says the White House approved a $9 billion request for Nvidia Blackwell chips to support AI use across spy agencies.
The talks come despite the Pentagon previously labeling Anthropic a "supply chain risk" after disputes over contract language tied to surveillance and autonomous weapons use.
The report also says the White House approved a $9 billion request for Nvidia Blackwell chips to support AI use across spy agencies.
๐จ๐ญ Global One SI - Prompt
$AMD/USD H2 #Indicator #Signal #Trading #GlobalOneAI
Here are some of Wall Street's price targets for $AMD stock
Baird $625
Evercore ISI $579
KeyBanc $530
Bernstein $525
Mizuho $515
Wells Fargo $505
Cantor Fitzgerald $500
Barclays $500
TD Cowen $500
Roth/MKM $500
Aletheia Capital $500
Baird $625
Evercore ISI $579
KeyBanc $530
Bernstein $525
Mizuho $515
Wells Fargo $505
Cantor Fitzgerald $500
Barclays $500
TD Cowen $500
Roth/MKM $500
Aletheia Capital $500
๐1
This is HUGE: OpenAl general purpose model had a breakthrough on famous 80 year old Erdos problem ๐คฏ!
โThis is believed to be the first time a general-purpose AI system has autonomously solved a major open problem central to a field of mathematics.โ
โThe breakthrough involves the โunit distance problem,โ originally posed by mathematician Paul Erdลs in 1946.โ
โFor decades, researchers believed square-grid-style constructions were essentially optimal for arranging points with the maximum number of equal distances.โ
โBut the AI discovered a completely new mathematical construction that produces more unit-distance pairs than previously thought possible.โ
โThe proof was independently checked by mathematicians and introduces new techniques in discrete geometry and combinatorics.โ
โThis is important because mathematics is one of the hardest domains for AI. A proof is either correct or wrong. There is no room for vague answers or benchmark tricks.โ
โThe result suggests future AI systems could help generate new discoveries in physics, algorithms, cryptography, biology, and other frontier sciences.โ
โThis is believed to be the first time a general-purpose AI system has autonomously solved a major open problem central to a field of mathematics.โ
โThe breakthrough involves the โunit distance problem,โ originally posed by mathematician Paul Erdลs in 1946.โ
โFor decades, researchers believed square-grid-style constructions were essentially optimal for arranging points with the maximum number of equal distances.โ
โBut the AI discovered a completely new mathematical construction that produces more unit-distance pairs than previously thought possible.โ
โThe proof was independently checked by mathematicians and introduces new techniques in discrete geometry and combinatorics.โ
โThis is important because mathematics is one of the hardest domains for AI. A proof is either correct or wrong. There is no room for vague answers or benchmark tricks.โ
โThe result suggests future AI systems could help generate new discoveries in physics, algorithms, cryptography, biology, and other frontier sciences.โ
๐1
Smart money is accumulating the infrastructure powering the next decade.
๐ Space
โ๏ธ Quantum
โ๏ธ NeoCloud
โก Energy
๐ง AI Infrastructure
These arenโt separate trends.
They fuel each other.
The physical layer of AI is becoming the biggest moat in the market.
Position early or pay a premium later.
๐ Space
โ๏ธ Quantum
โ๏ธ NeoCloud
โก Energy
๐ง AI Infrastructure
These arenโt separate trends.
They fuel each other.
The physical layer of AI is becoming the biggest moat in the market.
Position early or pay a premium later.
Forwarded from ๐ขWTI, Brent Crude Oil, NGAS Trading ๐
๐จTrump says a deal with Iran is "largely negotiated"
Iranian officials are talking about a 14-point framework.
Hormuz could reopen.
But nothing is signed.
Here's what's real and what isn't โ ๏ธ
Let's be precise about what we actually know.
What is confirmed:
โ Trump: deal is "largely negotiated," will include reopening Hormuz and lifting the US naval blockade
โ Iranian officials: a draft 14-point framework exists, providing a basis for further talks
โ Structure: end active hostilities
โ gradual Hormuz reopening
โ 30โ60 day negotiation window on nuclear issues and sanctions
Sources: Reuters, CNN, Axios, Al Jazeera, TIME all describe a draft framework, not a signed agreement
What is not confirmed:
โ Hormuz "fully reopened" on day one reporting says gradual, not immediate
โ No transit fees analyst speculation, no agreed text
โ Zero US payments drafts describe partial sanctions relief and frozen asset releases, which contradicts a clean "no payments" framing
โ Full lifting of Iranian oil sanctions phased relief tied to nuclear steps, not unconditional
โ Lebanon as formal ceasefire signatory likely included in de-escalation but not confirmed as a named party
โ Exactly "60 days" for nuclear talks 30โ60 day window reported, precise figure unconfirmed โก
This is high-probability direction of travel toward Hormuz reopening and partial sanctions relief.
It is not a done deal.
Oil and LNG prices will move on the framework rumour.
They would move hard in the opposite direction if the text falls apart ๐
A signed deal reopens Hormuz, brings back 12.8 mtpa of Qatari LNG capacity over years, and gradually unlocks Iranian barrels.
A collapsed deal keeps the current supply shock and potentially escalates it.
Both outcomes are still on the table.
Nothing is signed yet ๐
Iranian officials are talking about a 14-point framework.
Hormuz could reopen.
But nothing is signed.
Here's what's real and what isn't โ ๏ธ
Let's be precise about what we actually know.
What is confirmed:
โ Trump: deal is "largely negotiated," will include reopening Hormuz and lifting the US naval blockade
โ Iranian officials: a draft 14-point framework exists, providing a basis for further talks
โ Structure: end active hostilities
โ gradual Hormuz reopening
โ 30โ60 day negotiation window on nuclear issues and sanctions
Sources: Reuters, CNN, Axios, Al Jazeera, TIME all describe a draft framework, not a signed agreement
What is not confirmed:
โ Hormuz "fully reopened" on day one reporting says gradual, not immediate
โ No transit fees analyst speculation, no agreed text
โ Zero US payments drafts describe partial sanctions relief and frozen asset releases, which contradicts a clean "no payments" framing
โ Full lifting of Iranian oil sanctions phased relief tied to nuclear steps, not unconditional
โ Lebanon as formal ceasefire signatory likely included in de-escalation but not confirmed as a named party
โ Exactly "60 days" for nuclear talks 30โ60 day window reported, precise figure unconfirmed โก
This is high-probability direction of travel toward Hormuz reopening and partial sanctions relief.
It is not a done deal.
Oil and LNG prices will move on the framework rumour.
They would move hard in the opposite direction if the text falls apart ๐
A signed deal reopens Hormuz, brings back 12.8 mtpa of Qatari LNG capacity over years, and gradually unlocks Iranian barrels.
A collapsed deal keeps the current supply shock and potentially escalates it.
Both outcomes are still on the table.
Nothing is signed yet ๐
๐จ๐ญ Global One SI - Prompt
$GOOGL/USD H2 #Indicator #Signal #Trading #GlobalOneAI
This media is not supported in your browser
VIEW IN TELEGRAM
$GOOGL is now processing 3.2 QUADRILLION tokens per monthโฆ up from just 9.7 trillion in May โ24.
Sundar showed the AI flywheel hitting escape velocity.
Search, Gemini, Cloudโฆ every surface is exploding with real usage.
This is parabolic demand at planetary scale.
The market still hasnโt fully priced it in.
$GOOGL remains one of the strongest asymmetric bets in Big Tech.
Sundar showed the AI flywheel hitting escape velocity.
Search, Gemini, Cloudโฆ every surface is exploding with real usage.
This is parabolic demand at planetary scale.
The market still hasnโt fully priced it in.
$GOOGL remains one of the strongest asymmetric bets in Big Tech.