Aggregator

US, Saudi Officials Confirm Kingdom's East-West Pipeline Attacked By Drones Out Of Iraq Summary

• Saudi oil pipeline attacked: US officials confirmed strikes on Saudi pumping stations, with drones likely launch from Iraq.
• Houthis expand in Yemen: Houthi forces are gaining control along the Red Sea coast, threatening key shipping routes.
• Oil and diesel prices rising: Russia's damaged refining capacity & shipping disruptions are also tightening fuel markets.
• Stagflation risk grows: Higher energy and shipping costs could push inflation higher while weakening global economic growth.

//--> //--> //--> Strait of Hormuz traffic returns to normal by November 30?
Yes 13% · No 88%
View full market & trade on Polymarket US Officials Confirm East-West Pipeline Was Attacked, Saudis Blame Iraqi militants

Open source satellite imaging has shown the crucial 'Hormuz bypass' East-West oil pipeline that cuts straight across Saudi Arabia has been on fire. A some 80km to 100km giant smoke plume has been observed. The damaged section is said to be located near the town of Al Mesba'ah.

• SAUDI BLAMES OIL PIPELINE ATTACK ON DRONES FROM IRAQ: AP
• IRAQ PM ORDERS PROBE INTO ATTACKS ON SAUDI ARABIA: STATEMENT

Amid speculation that this is probably the result of a major Houthi attack out of Yemen, CNN in a new Friday report cites two US officials who've given confirmation that the pipeline was struck by projectiles on Thursday. Iraq was named as possible attack origin point. According to the report:

An early analysis found that pump stations, which are located next to the pipeline itself, were hit, one of the US officials said. A satellite image taken Friday appears to show extensive fire damage at one pumping station, and an image of a different pumping station, taken Thursday, showed a small fire sending up plumes of thick, black smoke.

It could have been the result of a drone attack by paramilitaries operating out of Iraq. "It was not immediately clear who was responsible for the strikes or if sections of the pipeline itself were damaged but one of the officials said it was struck by drones originating from Iraq," CNN reports. "It was also not immediately clear how long it would take to repair the damage, sources said."

On Sept. 10, via EUMETSAT/CNN

News on Saudi oil has gone from worse to worse, as in Yemen the Houthis have reportedly taken over the entire Red Sea coastline. They have renewed the threats to attack all Saudi (as well as Israeli) shipping, but have also sought to assure transit is 'safe' for other international vessels.

The pumping station struck yesterday is located near the town of Al Mesba'ah.

According to one of the U.S. officials who spoke to CNN, the drones that struck the pumping station along Saudi Arabia’s East-West pipeline originated from Iraq. It remains unclear whether the pipeline… pic.twitter.com/puOWM5VRlX

— OilPrice.com (@OilandEnergy) September 11, 2026 Threats Rapidly Converge

Three converging threats to watch are intensifying pressure on global energy markets: damage to Russian refining capacity, ongoing disruption at Hormuz, and new, expanding Houthi threats to Red Sea shipping. The resulting physical market squeeze extends well beyond the Gulf area. Record-high diesel prices in the US and other markets, such as China's return to buying crude, raise the risk that oil markets will remain exceptionally tight into the Northern Hemisphere winter. 

What we know so far is that the Russia-Ukraine war has knocked out a whole bunch of refining capacity and halted exports of critical fuels from Russia. The Gulf area has seen an escalation in fighting this week as the Hormuz chokepoint remains open with tanker transits but still limited and far from pre-war levels. 

Couple this all with the chokepoint madness still being disrupted and new developments overnight: Iran-backed Houthi forces advanced toward a strategic port near the southern entrance to the Red Sea, threatening to tighten control on a second critical shipping corridor. 

Houthis Take Yemen's Red Sea Coast

Bloomberg reported that Houthi rebels have gained ground around Mokha, with some geopolitical analysts reporting that the Yemeni port city has fallen. 

🇾🇪 Total collapse: The entire coastline from Dhubab to Mayyun island was abandoned. Mayyun, Duhbab and Murd now controlled by Houthi forces as of September 11

A major global event. The Houthis now have the capacity to block ships now more than ever. Bab Al Mandab at their mercy https://t.co/ycUtzCcPq8 pic.twitter.com/EfVc5GdbZv

— The Cube (@war_cube) September 11, 2026

Its fall would give the Houthis another coastal stronghold alongside Hodeida and the ability to control more of the Bab el-Mandeb Strait, suggesting commercial traffic could begin to drop and transits would be rerouted around the Cape of Good Hope, increasing shipping time and freight costs. 

Alternative Route: Cape of Good Hope

On top of this, China has returned to global oil markets, and ex-Goldman Commodities head Jeff Currie warned Thursday that this is the real driver of soaring crude prices. 

Saudi Key East-West Pipeline on Fire

Also overnight, Saudi Arabia's East-West oil pipeline appears to have been struck by Houthi forces, which the pipeline served as an oil escape route, effectively bypassing the Hormuz chokepoint to the Red Sea.

Andrew Farrand, a political-risk analyst at Horizon Engage, described the rapid Houthi advance as a major setback for Saudi efforts in Yemen, warning that it could bring the Houthis closer to territory overlooking the waterway's narrowest section. 

Simultaneous disruption of Bab el-Mandeb Strait and Strait of Hormuz creates a two-sided squeeze: less energy can leave the Gulf, while tankers that can move face longer, more expensive journeys. It also threatens Saudi Arabia's Red Sea terminal.

Beyond Hormuz and Bab el-Mandeb, these are the main straits to watch: 

Global Maritime Chokepoints

Where the routes narrow

What could interrupt commercial traffic

Stage Set for Stagflationary Squeeze?

The twin disruptions threaten shipping corridors that carried roughly a quarter of global seaborne oil trade through Hormuz and through Bab el-Mandeb, while jeopardizing a Red Sea shipping route central to trade between Asia and Europe and risking ignition of a stagflationary squeeze as diesel prices soar to new highs that eventually feed into supply chains, freight costs, and ultimately, at a lag, higher prices on store shelves. 

Tyler Durden Fri, 09/11/2026 - 12:53

Now They Think Babies Can Be Racist...

Authored by Steve Watson via Modernity News,

A woman sat in front of a camera this week and told the internet she had witnessed a "racist baby"...

Not a joke account. Not a sketch. A straight-faced lecture about an infant who looked at someone for a few seconds - and, in her telling, revealed the original sin of race hatred barely out of the womb.

The woman declares that "watching children mirror racist behavior is WILD." She describes a baby staring at her 'friend of colour', the mother looking stricken, and then announces that the child is "literally mimicking" hostility toward "black and brown people."

This retard says she saw a "racist baby"

This person is a legit retard.. this is somehow not a joke.

Unbelievable pic.twitter.com/R5UUfig5AN

— Retard Finder (@IfindRetards) September 10, 2026

She also flashes a 2012 Science Daily headline as if it were a conviction.

This is the new priesthood. A baby looks at a black person and the verdict is racism.

Anyone who has any experience at all with babies knows they stare. At glasses. At beards. At bald heads, bright coats, dogs, ceiling fans, and their own hands. They are cataloguing the world with amazement, not forming a manifesto.

Obviously, she's never had children. Babies also stare at people with glasses if they've never seen a person with glasses, they stare at all kinds of things they've never seen before. It's called learning and it's up to the parents to teach them from then on.

— Dee Dee (@littleDisme) September 10, 2026

The woman in the video treats curiosity as contamination. She reaches for a University of Massachusetts Amherst study from 2012 - "Infants begin to learn about race in the first year" - and reads it as proof that a pre-verbal child is already a bigot.

The paper does not say that. It found that by nine months, babies get better at recognising familiar-looking faces and worse at telling apart unfamiliar ones.

Psychologist Lisa Scott, an author of the work, said the results "suggest that biases in face recognition and perception begin in preverbal infants, well before concepts about race are formed."

Face recognition is not a worldview. A nine-month-old does not have a theory of systemic oppression.

They diagnosed a baby with racism. That's the priesthood now.

— Storm Files (@TheStormFiles) September 10, 2026

Leftwing women hate babies so much, they can't hide it.

— Ames (@VivaLaAmes11) September 10, 2026

This is not an isolated freak-out. It sits inside a wider epidemic of deranged white leftist women.

They police stares, invent motives for infants, and treat ordinary childhood as a moral emergency. These people staff HR departments, classrooms, nurseries and social-work offices. They write the guidance.

They decide which toddler is a "perpetrator." exporting their own neurosis onto the smallest most innocent people in the room.

Her assumptions she has about a babies thoughts on black people are the thoughts she has about black people.

It's a clever way to say you hate black people.

— Mike Mossey (@Mikemossey) September 10, 2026

In fairness, the baby refused to acknowledge his privilege, denounce systemic oppression, or even apologize for it's unconscious bias.

The baby needs Infant Bias Intervention, Diapers, Diversity & Inclusion or Critical Cradle Theory as soon as he finishes the bottle.?

— SUPA DUPA FUPA™ (@TheFupaverse) September 10, 2026

You are a gift and a savior to black people everywhere I don't know what they would do without you

— Trip Pache (@DrSlayyyy) September 10, 2026

pic.twitter.com/Rpfrq2e6xv

— The Vat of Butter (@theVatOfButter) September 10, 2026

The viral clip would be easier to laugh off if governments were not writing the same fantasy into official guidance.

In Wales, childcare workers have been trained under a taxpayer-funded programme - more than £1.3 million via the Welsh Government - to spot "racist incidents" among toddlers and, if staff decide the behaviour could amount to a hate crime, contact police.

The toolkit from Diversity and Anti-Racist Professional Learning (DARPL) at Cardiff Metropolitan University has been circulated to more than 300 nurseries, playgroups and childminders.

Staff are told to audit toys and books for "diversity," discuss skin colour with very young children, and rate their own grasp of "white privilege" on a one-to-five scale. If the incident falls short of a hate crime, workers are advised to offer "age-appropriate learning support opportunities for the perpetrator." If that is "met with resistance," a disciplinary route appears on a flowchart. The "perpetrator" may still be in nappies.

Lucy Marsh, communications officer at the Family Education Trust, asked "Do they want toddlers to have a criminal record?"

Kent Police went one better. A one-year-old girl was logged as a crime suspect after allegedly causing a minor injury to another toddler.

Freedom of Information figures showed 683 children under 10 reported for offences over three years - including six two-year-olds, 11 three-year-olds and 20 four-year-olds.

None can be prosecuted. The age of criminal responsibility in England and Wales is 10. Kent County Council cabinet member for children's services, Councillor Paul Webb, called the numbers "not great." Chief Superintendent Rob Marsh said the focus was "safeguarding rather than punishment." The record still exists. A baby is on a crime log.

The contempt is not limited to infants who look at people the wrong way. Wanting people to have children at all has been recast as an extremist project. Politico framed a natalism conference as proof that "the far right is so obsessed with making babies, they just held a whole conference about it."

Reporter Gaby Del Valle treated collapsing fertility as a punchline and smeared those who notice it as plotters of "a total social overhaul."

Meanwhile the CDC recorded a U.S. fertility rate of 1.6 in 2023 against a replacement level of 2.1. A Lancet study put the global rate at 2.23 and warned the world's population will start to fall within decades. Co-author Dr Natalia Bhattacharjee said declining fertility "will completely reconfigure the global economy and the international balance of power." That is biology, not a rally chant.

So the sequence is complete. Have children and you are suspect. The children themselves are suspect. A glance is evidence. A nursery becomes a reporting desk. The state writes "racist" next to a toddler and calls it safeguarding.

A baby is not a political actor. It does not owe a land acknowledgement between feeds. It does not need Critical Cradle Theory. It needs parents who are attached to reality and institutions that stop treating childhood as a crime scene.

Tyler Durden Fri, 09/11/2026 - 12:45

Here come the frogs.

This episode packs a wild mix of high-stakes crime and flat-out bizarre headlines. We break down TMZ’s investigation into a hidden fifth plane on 9/11, crucial updates in the murder trial for Charlie Kirk’s shooter and major developments in the Hayden Panettiere death investigation as federal agents step in. The chaos only ramps up from...

Mayor Zohran Mamdani and Rep. Alexandria Ocasio-Cortez were seen yukking it up during the somber reading of 9/11 victims at Ground Zero during the 25th anniversary service.

Your Post experts picked the entire NFL slate this week and every week.

"Omg I remember when you first had them," one fan commented on the pop icon's sweet upload.

"Omg I remember when you first had them," one fan commented on the pop icon's sweet upload.

WHO has said the outbreak remains out of control and is on track to surpass the 2014-2016 West Africa Ebola outbreak.

The unregulated sale of the weed drinks in grocery stores and bodegas is happening despite New York law clearly requiring a specific license to sell them.

The 15-year-old son of Kim Zolciak and Kroy Biermann appeared in court for a bond hearing Friday.

The 15-year-old son of Kim Zolciak and Kroy Biermann appeared in court for a bond hearing Friday.

Here's the 411.

Border Patrol agents arrested Indian national Rohit Rohit, 22, as he attempted to cross into the US over the Arizona border in August 2023. Under the rules of the Biden Administration, Rohit was released a month later under an “order of recognizance.” jQs5tQN3

The college football bettors guide is back with a new look this year. Seven contestants, two winners in The Post’s standings and best bets standings. This year’s picks have a new look and new information to delve into.  In Week 1, The Post’s Howie Kussoy got off to a strong start alongside Ethan Sears, who...

DeepSeek's New Hyper-Efficient Model Stokes Fears Over Korea's Memory Makers

Samsung Electronics and SK Hynix each fell more than 3% in Seoul on Friday after DeepSeek said its newest AI model needs a fraction of the memory required by its predecessor. Both stocks had been trying to recover from July's selloff and remain more than 25% below their highs. Local retail traders, who helped drive the rally earlier this year, have sold around $10 billion of the pair this month alone.

The trigger came Thursday out of Hangzhou. DeepSeek's V4.1-Flash is fast, cheap and, by the company's own numbers, stronger than its flagship model. But the title of DeepSeek's paper had nothing to do with intelligence benchmarks. It called the release "Pushing the Limits of KV Cache Compression," and its abstract identifies memory consumption during long AI sessions as the main obstacle to making these models cheaper to run.

What DeepSeek Shipped

V4.1-Flash is a 552-billion-parameter model (parameters are the numerical values a model learns during training). DeepSeek does not activate all 552 billion for every word. It uses 8 billion parameters while processing input and 16 billion while generating output, a design that lowers the amount of computing needed for each step.

The model can handle a context window of one million tokens, meaning roughly a million small pieces of text or other input can remain available to it during a session. It also reads images natively, and its model weights are available under an MIT license, allowing anyone with sufficient hardware to run it. Nine providers were serving it through OpenRouter within a day of release.

The memory problem needs a little more explanation. As an AI model works through a long document, conversation or agent task, it keeps a running record of what it has already processed so it does not have to recalculate everything from scratch each time it generates another token. That record is called the key-value cache, or KV cache.

The cache is normally stored in high-bandwidth memory, or HBM. HBM consists of stacks of DRAM placed next to the processor to move data extremely quickly. It is among the fastest and most expensive memory in production, and booming AI demand for it helped turn SK Hynix, Samsung and Micron into some of the biggest semiconductor trades of 2026. As an AI session gets longer, however, the cache keeps growing. In long-running agent workloads, the memory needed for the cache can eventually exceed the memory occupied by the model itself.

DeepSeek has attacked that problem directly. Its new architecture reuses cached information across layers of the network, stores the cache at 4-bit precision and reconstructs part of it when needed rather than storing the entire record. The result, according to DeepSeek, is 890 bytes of cache per token, one-quarter of what the previous Flash model kept in HBM and one-eighth of what it wrote to SSD storage.

DeepSeek's post on X gave the commercial reason plainly. Cached input can account for a large share of the cost of running an AI agent. Shrink the cache, and the bill falls with it.

Memory Per Token, Down 437-Fold

The figure in DeepSeek's paper that matters most for Seoul is the second one. It tracks the amount of cache memory required for every token across successive generations of DeepSeek models. From the company's first release in January 2024 to V4.1-Flash, that figure has fallen 437-fold.

KV cache per token, indexed to the V4 predecessor at full context. Chained from DeepSeek's stated ratios; V4.1-Flash's persistent SSD cache falls further, to one-eighth of V4-Flash. Source: DeepSeek model card; Wccftech.

The decline did not begin this week. April's V4 had already cut KV-cache requirements to one-tenth of its predecessor at the full context window. September's model reduced them by another three-quarters. DeepSeek describes V4.1-Flash as the smallest member of a new architecture family designed to scale to larger models.

On the other side of that chart are the growth assumptions embedded in the memory trade. Micron says its entire calendar 2026 HBM supply is already contracted on both price and volume. In December, the company told investors the HBM market could grow from about $35 billion in 2025 to roughly $100 billion in 2028. By June it had moved the $100 billion estimate forward to 2027.

Micron reports earnings on September 30 and has guided the quarter to roughly $50 billion in revenue, about 350% above the year-earlier period, with an 86% gross margin.

How much premium memory each unit of AI work requires is the whole point. 

Out With The 'Old' 

This seems a little risky, but beginning September 14, DeepSeek will route every request for V4-Pro, its 1.6-trillion-parameter flagship, to V4.1-Flash and charge the lower Flash rate until a V4.1-Pro arrives. No date has been given for the larger model.

DeepSeek says the smaller model now beats V4-Pro on performance, cost, speed and the time required to complete a task. Model size alone is becoming a worse guide to how much computing and memory a useful AI system will consume.

Against the leading American models, V4.1-Flash fits the pattern that has held for much of this year. On DeepSeek's own benchmark tables, it narrowly beats the best reported score from Claude Opus 5 or GPT-5.6 Sol on Terminal-Bench 2.1, DeepSWE, CyberGym and Humanity's Last Exam when tools are allowed.

On newer and harder tests, however, the American frontier remains well ahead. V4.1-Flash scores 30.0 against Opus's 43.3 on Terminal-Bench 3.0, 31.2 against 51.8 on version 4.0, 20.3 against 37.0 on ProgramBench, and 15.3 against GPT-5.6 Sol's 33.7 on ExploitGym. The results have not been independently verified.

DeepSeek's open model is clearing benchmarks that defined the frontier last year at a fraction of the price, while the American frontier keeps moving to harder tests. It is the same pattern seen with V4-Flash earlier this year.

86x Cheaper!?

According to X user NIK (@ns123abc), V4.1-Flash is 86 times cheaper than the American flagships. This is because AI providers charge separately for input tokens, the material sent into the model; output tokens, the material the model generates; and cached input, previously processed material that can be reused without running the full computation again.

At peak hours, V4.1-Flash costs $0.30 per million input tokens and $1.20 per million output tokens, with prices cut in half during off-peak hours. Anthropic charges $5 and $25 for Claude Opus 5. On output, that makes DeepSeek roughly 20 times cheaper at peak and about 40 times cheaper off-peak.

Deepseek just dropped v4.1 flash, fully open weights

it beats gpt 5.6 sol, opus 5 and every chinese model on coding and cybersecurity at ~86x cheaper cost per million tokens running at 420-507 tok/s = faster than gemini 3.8 flash

"smallest model in our new architecture family"… pic.twitter.com/02Fft2OKTB

— NIK (@ns123abc) September 10, 2026

The 86-fold figure comes from cached input. Anthropic charges 50 cents per million cached tokens, while DeepSeek charges six-tenths of a cent. Cached input is precisely the cost DeepSeek's new memory architecture was built to reduce.

Peak-hour list prices per million tokens, log scale. The 83x on cached input is what the listed rates give; the 86x circulating on X is that post's own math. Source: VentureBeat; Dataconomy.

Meanwhile, DeepSeek raised its prices only a month ago. On August 16 it moved V4-Flash from flat rates of $0.14 for input and $0.28 for output to peak rates of $0.44 and $1.32. Thursday's release brought those prices back down, although not to July's levels, while giving V4-Pro customers a reduction of roughly 70%.

V4-Flash list prices at launch, after the August 16 hike, and at Thursday's V4.1-Flash release. Off-peak rates are half of peak. Source: TechJack Solutions; VentureBeat.

The timing also comes as DeepSeek moves toward the public markets. On Wednesday, Reuters reported that the company had hired CITIC Securities to prepare for a Shanghai listing. That followed a June financing round of about $7.4 billion involving investors including Tencent and CATL, along with reports of another potential round at a valuation near 500 billion yuan.

July Was Supply, September Is Demand

The July selloff in Korean memory stocks centered on supply. SK Hynix signaled a major increase in spending, while Chinese memory manufacturers continued ramping cheaper output.

Friday's scare came from the other side of the market: each unit of AI activity needing fewer memory chips.

Micron and SanDisk, which held up overnight, were rising in Friday's premarket after Oracle's cloud results. The Korean names, where leverage and local retail participation are greater, took the immediate hit.

Moves of 5% or more in a single day remain common for both Samsung and SK Hynix. Bloomberg notes that volatility remains near levels last seen during the 2008 financial crisis and the Covid shock. The shares also look inexpensive on conventional measures. Samsung trades at about 2.7 times book value and Hynix at 5 times, compared with roughly 11 times for the Philadelphia Semiconductor Index. Both Korean companies trade near 4 times forward earnings, versus 19 times for the index.

Within hours, Fibonacci Asset Management's Jung In Yun said cheaper AI could drive more usage and offset the efficiency gains. Eugene Asset Management's Ha SeokKeun called the issue a near-term concern.

When Seoul reopened after the Lunar New Year on January 31 last year, eleven days after DeepSeek's R1 release, SK Hynix fell as much as 12% in a day and Samsung dropped 4%. Investors initially feared that more efficient models would weaken demand for AI hardware. Instead, AI spending kept climbing, and the argument that lower costs stimulate greater usage won the year that followed.

R1 challenged the amount of computing required to produce useful AI. V4.1-Flash is attacking the amount of memory required for every token, and DeepSeek has published a chart, running back to January 2024, showing that requirement falling by more than 400-fold. The company also says the same architecture is intended for larger models.

Export Controls And Huawei's Own HBM

Since December 2024, U.S. export controls have barred sales of advanced HBM to China, making access to fast memory one of the hardware constraints on Chinese AI developers.

Huawei has been working on a domestic alternative. Its Ascend 950DT uses Huawei-made HiZQ 2.0 memory, with 144 gigabytes of capacity and bandwidth of 4 terabytes per second. That remains well behind the HBM SK Hynix supplies for Nvidia's leading accelerators.

DeepSeek's V4 in April was the first frontier model validated on Huawei Ascend hardware alongside Nvidia chips. The company also has a reported order for 160,000 Ascend 950DT processors for a gigawatt-scale data center in Ulanqab.

So - a model that needs one-quarter as much memory per token can be deployed more broadly on hardware that has less memory to offer. The V4.1-Flash model card does not identify the hardware used to train the model.

DeepSeek has not announced a date for the larger member of the V4.1 family. Micron reports on September 30.

Tyler Durden Fri, 09/11/2026 - 12:25

CCTV footage shows the train travelling along the platform, dragging the helpless passenger towards the tracks as he struggles to free himself.

The Jets’ kicking mess has been resolved … again … for now.

Iran-backed Houthi rebels in Yemen tried to use Anthropic’s powerful Claude AI model to craft software for use in ballistic missiles, according to an alarming report released by the company.

Anthropic says it caught and shut down large-scale attempts by Chinese AI labs including Alibaba, Moonshot and DeepSeek to use Claude to train their own AI models