This media is not supported in your browser
VIEW IN TELEGRAM
Anduril is reportedly expanding its Arsenal-1 facility in Ohio with a $1B investment to scale production of defense systems.
The site is expected to cover ~464,000 m² and employ over 4,000 people in the coming years.
#USA #Anduril #Defense #Production
The site is expected to cover ~464,000 m² and employ over 4,000 people in the coming years.
#USA #Anduril #Defense #Production
🇨🇭 Global One SI - Prompt
Anduril is reportedly expanding its Arsenal-1 facility in Ohio with a $1B investment to scale production of defense systems. The site is expected to cover ~464,000 m² and employ over 4,000 people in the coming years. #USA #Anduril #Defense #Production
Company officials report a ~15× increase in output between 2025–2026, highlighting growing demand for unmanned and defense technologies at scale.
(Source: Jane’s Defence)
#MilitaryTech #Industry #Aerospace
(Source: Jane’s Defence)
#MilitaryTech #Industry #Aerospace
🇨🇭 Global One SI - Prompt
The UK has unveiled a new long-range strike drone TigerShark, developed by MGi Engineering and first tested in April 2026. Still in early testing, the drone is described as a deep-strike platform, closer to a cruise missile-type system than a conventional…
The system reportedly reaches ~750 km/h, carries up to 300 kg, and may operate over 1,000 km with GPS/INS navigation and terrain-referencing.
#Defense #Autonomous #Aerospace
#Defense #Autonomous #Aerospace
Forwarded from 🛢WTI, Brent Crude Oil, NGAS Trading 📊
Saudi Arabia just did something it has never done before.
$19.5 premium per barrel over benchmark
Look at that chart.
Every bar since 2021 between $0 and $10.
Then 2026.
Straight to $19.5, a vertical spike unlike anything in the data.
This is what a seller's market looks like when the seller controls the only bypass route.
Think about what Saudi Arabia is saying:
Hormuz is closed✅
Every bypass pipeline is maxed ✅
Asia is desperate for our quality barrels ✅
You have nowhere else to go✅
So pay $19.5 over benchmark or go without.
Asia will pay.
Because the alternative empty refineries, shuttered factories, social unrest costs infinitely more.
This premium lands directly on:
⛽ Every petrol station in Asia
✈️ Every airline ticket price
🏭 Every factory input cost
$19.5 premium per barrel over benchmark
Look at that chart.
Every bar since 2021 between $0 and $10.
Then 2026.
Straight to $19.5, a vertical spike unlike anything in the data.
This is what a seller's market looks like when the seller controls the only bypass route.
Think about what Saudi Arabia is saying:
Hormuz is closed✅
Every bypass pipeline is maxed ✅
Asia is desperate for our quality barrels ✅
You have nowhere else to go✅
So pay $19.5 over benchmark or go without.
Asia will pay.
Because the alternative empty refineries, shuttered factories, social unrest costs infinitely more.
This premium lands directly on:
⛽ Every petrol station in Asia
✈️ Every airline ticket price
🏭 Every factory input cost
WSJ just published a piece. OpenAI and Anthropic just showed that modern AI is becoming a race to finance compute, not just invent better models.
Their internal forecasts say the hardest part is no longer selling AI, but paying for the giant clusters of chips needed to train each new model and run inference for every user prompt.
OpenAI expects $121B in compute spending for 2028 and an $85B burn that year even with much higher sales, while Anthropic projects the same pattern at a smaller scale.
Both companies therefore present 2 profit views: one that removes research compute and makes the business look close to viable, and one that includes it and shows how far they still are from normal software economics.
Their internal forecasts say the hardest part is no longer selling AI, but paying for the giant clusters of chips needed to train each new model and run inference for every user prompt.
OpenAI expects $121B in compute spending for 2028 and an $85B burn that year even with much higher sales, while Anthropic projects the same pattern at a smaller scale.
Both companies therefore present 2 profit views: one that removes research compute and makes the business look close to viable, and one that includes it and shows how far they still are from normal software economics.
👀1
$SIVE $SOI $AEHR 🔥
The market is still pricing in 2024/2025 revenues while the real volume ramp kicks off in H2 2026. LightCounting is already forecasting a $110B+ TAM with explosive growth - and the overlapping S-curves (800G → LPO/LRO → CPO → 1.6T/3.2T) are setting up a true multi-year supercycle.
The pure-play names are positioned exceptionally well:
$SIVE → CW-Laser + ELS technology - direct winner from the next big architecture shift in CPO/SiPH
$SOI → Silicon-on-Insulator substrates & photonics foundry - absolutely critical for the new generation
$AEHR → Test equipment that prints money once yield challenges hit and volumes really scale
These three are still not properly priced in. Forward P/E compression + massive TAM expansion = huge re-rating upside starting H2 2026.
We’re at the very beginning of the next major inflection point. Anyone holding the cleanest photonics names right now is sitting in the perfect seat.
NFA - but the asymmetry is absolutely brutal.
The market is still pricing in 2024/2025 revenues while the real volume ramp kicks off in H2 2026. LightCounting is already forecasting a $110B+ TAM with explosive growth - and the overlapping S-curves (800G → LPO/LRO → CPO → 1.6T/3.2T) are setting up a true multi-year supercycle.
The pure-play names are positioned exceptionally well:
$SIVE → CW-Laser + ELS technology - direct winner from the next big architecture shift in CPO/SiPH
$SOI → Silicon-on-Insulator substrates & photonics foundry - absolutely critical for the new generation
$AEHR → Test equipment that prints money once yield challenges hit and volumes really scale
These three are still not properly priced in. Forward P/E compression + massive TAM expansion = huge re-rating upside starting H2 2026.
We’re at the very beginning of the next major inflection point. Anyone holding the cleanest photonics names right now is sitting in the perfect seat.
NFA - but the asymmetry is absolutely brutal.
The entire Hollywood assistants are making AI part of the creative pipeline.
However, this is not an AI takeover story so much as a labor squeeze story.
Because just as with many other industries in Hollywood as well, pressure is rising to produce more in a shorter time.
"Confronted with larger workloads and a shrinking headcount, AI — both the kind officially approved by companies and more surreptitious uses — has made its way via support staff into essential Hollywood workflows, including the creative development process."
However, this is not an AI takeover story so much as a labor squeeze story.
Because just as with many other industries in Hollywood as well, pressure is rising to produce more in a shorter time.
"Confronted with larger workloads and a shrinking headcount, AI — both the kind officially approved by companies and more surreptitious uses — has made its way via support staff into essential Hollywood workflows, including the creative development process."
This media is not supported in your browser
VIEW IN TELEGRAM
OpenClaw alternative.
PicoClaw (by Chinese AI company Sipeed with 27K+ Gihub stars ⭐️).
LLM orchestration, multi-channel messaging, tools/skills, MCP support while running on ~$10 single-board computers with a core footprint of <10 MB RAM.
picoclaw.io
PicoClaw (by Chinese AI company Sipeed with 27K+ Gihub stars ⭐️).
LLM orchestration, multi-channel messaging, tools/skills, MCP support while running on ~$10 single-board computers with a core footprint of <10 MB RAM.
picoclaw.io
✍1
This media is not supported in your browser
VIEW IN TELEGRAM
Jaguar Land Rover Tests Elios 3 #Drone, Slashes Inspection Time by 95%