Toggle light / dark theme

Get the latest international news and world events from around the world.

Log in for authorized contributors

Skull channels as gateways for immune surveillance of the brain

The central nervous system (CNS), once considered immune privileged, is, in fact, protected by a dynamic immune compartment located in the meninges. These immune cells are crucial for brain protection, controlling neuroinflammation, and also supporting neural function. Recent studies show that meningeal immune cells, including myeloid and B cells, partly originate from the skull bone marrow, which communicates directly with the meninges via specialized skull channels identified in mice and humans. These channels allow immune cell trafficking and cerebrospinal fluid (CSF) exchange, positioning the skull marrow as a CSF-sensing niche. Therefore, understanding skull–meninges–brain interactions reveals a functional connection between bone marrow and the CNS, offering new insights into neuroimmune regulation and potential therapeutic strategies in neuroinflammatory conditions.

Galactic Gems Glisten in New Gallery From NASA’s Chandra

This gallery displays a collection of 16 galactic images from Chandra and other telescopes. Astronomers put galaxies into three main categories: spirals like our own Milky Way, ellipticals that are older and likely the results of mergers, and irregulars that can encompass a wide range of galactic phenomena. All types are represented in this collection. Each galactic image contains X-ray data from Chandra combined with data from telescopes such as NASA’s Webb, Hubble, IXPE, Swift, and NuSTAR, and others both in space and on the ground.

Nvidia ACIE Revenue Hit $40 Billion. Memory Costs Are Already Squeezing It

Nvidia’s ACIE segment, AI Clouds, Industrial, and Enterprise, hit $40.3 billion in Q2 FY27, up 138% year over year and now representing nearly half of data center revenue. The same earnings call carried a quieter warning: memory scarcity is pushing Nvidia’s own costs higher and could weigh on gross margins in coming quarters. The growth is real. So is the cost pressure sitting underneath it.

Nvidia ACIE revenue just delivered the clearest proof yet that industrial and enterprise AI spending is real money, not a hyperscaler side story. Nvidia’s AI Clouds, Industrial, and Enterprise category generated $40.3 billion in the quarter ended July 2026, up 138% year over year and 25% sequentially, according to Fortune’s coverage of the earnings report. CFO Colette Kress said non-hyperscaler growth, spanning sovereign regional NeoClouds, enterprise edge, and air-gapped data centers, now represents roughly half of Nvidia’s data center business.

Nvidia created the ACIE category specifically to separate hyperscaler cloud spending from AI-native clouds, sovereign AI, and on-premises enterprise and industrial demand, according to Yahoo Finance’s earnings call highlights. That reporting change matters on its own: Nvidia ACIE revenue needed its own line item because industrial and enterprise AI demand had grown large enough to distort the picture if it stayed folded into the broader hyperscaler number. Data center revenue overall reached $89 billion for the quarter, with hyperscale customers contributing $48.7 billion and ACIE the remaining $40.3 billion.

Physicists Just Directly Imaged The Quantum Fluctuations of Empty Space

Empty space is never actually truly empty.

Even a quantum field in its lowest-energy state, the vacuum, has unavoidable fluctuations that arise from the Heisenberg uncertainty principle: Certain pairs of properties cannot both have precisely defined values at the same time.

In real-world terms, this means that a quantum field should resemble a staticky TV, and plenty of indirect evidence suggests this is the case.

A Tougher Extracellular Matrix Strengthens Tendons in Rats

Researchers have found that an upstream promoter of two extracellular matrix proteins increases healing and strength capabilities in the tendons of rats.

Tendon injuries in older people

Previous research has found that, like with many other injuries, tendon injuries have an age-related component. The tendons of the biceps and rotator cuffs are more commonly injured in older people than in younger people [1], and older people heal slower after forearm injuries and have less range of motion in the fingers for a longer time [2].

Going beyond simple models of neural networks: Extended mean-field theory offers a better approach

Physics is most readily applied to relatively simple systems: a pendulum, two electrons colliding or the structure of the solar system. But when systems become complicated—when many particles interact with one another, in condensed matter systems such as gases and fluids or in the cosmology of the early universe—simplifications can be made using a technique called classical or extended mean-field theory.

A research group from Princeton University, led by lead author Luca Di Carlo, has applied extended mean-field theory to networks of biological neurons. Their work is published in the journal Physical Review Letters. They found that the simplest versions of mean-field theory failed to accurately describe the activity of these networks, but an enhanced version was able to do so.

/* */