Toggle light / dark theme

It’s obvious when a dog has been poorly trained. It doesn’t respond properly to commands. It pushes boundaries and behaves unpredictably. The same is true with a poorly trained artificial intelligence (AI) model. Only with AI, it’s not always easy to identify what went wrong with the training.

Research scientists globally are working with a variety of AI models that have been trained on experimental and theoretical data. The goal: to predict a material’s properties before taking the time and expense to create and test it. They are using AI to design better medicines and industrial chemicals in a fraction of the time it takes for experimental trial and error.

But how can they trust the answers that AI models provide? It’s not just an academic question. Millions of investment dollars can ride on whether AI model predictions are reliable.

The resuspension of seafloor sediments—triggered by human activities such as bottom trawling as well as natural processes like storms and tides—can significantly increase the release of carbon dioxide (CO2) into the atmosphere. When these sediments are exposed to oxygen-rich seawater, large-scale oxidation of pyrite occurs.

This reaction plays a much greater role in CO2 emissions than previously assumed, exceeding the contribution from the of . A new study, published in Communications Earth & Environment, provides the first quantitative evidence of this effect in the western Baltic Sea.

“Fine-grained, muddy sediments are important reservoirs of organic carbon and pyrite,” says lead author Habeeb Thanveer Kalapurakkal, a Ph.D. student in the Benthic Biogeochemistry working group at GEOMAR.

A study conducted by CNRS researchers describes a new method of recycling silicone waste (caulk, sealants, gels, adhesives, cosmetics, etc.). It has the potential to significantly reduce the sector’s environmental impacts.

This is the first universal recycling process that brings any type of used silicone material back to an earlier state in its where each molecule has only one silicon atom. And there is no need for the currently used to design new silicones. Moreover, since it is chemical and not mechanical recycling, the reuse of the material can be carried out infinitely.

The associated study is published in Science.

To learn more about the nature of matter, energy, space, and time, physicists smash high-energy particles together in large accelerator machines, creating sprays of millions of particles per second of a variety of masses and speeds. The collisions may also produce entirely new particles not predicted by the standard model, the prevailing theory of fundamental particles and forces in our universe. Plans are underway to create more powerful particle accelerators, whose collisions will unleash even larger subatomic storms. How will researchers sift through the chaos?

The answer may lie in . Researchers from the U.S. Department of Energy’s Fermi National Accelerator Laboratory (Fermilab), Caltech, NASA’s Jet Propulsion Laboratory (which is managed by Caltech), and other collaborating institutions have developed a novel high-energy particle detection instrumentation approach that leverages the power of quantum sensors—devices capable of precisely detecting single particles.

“In the next 20 to 30 years, we will see a in particle colliders as they become more powerful in energy and intensity,” says Maria Spiropulu, the Shang-Yi Ch’en Professor of Physics at Caltech.

Could the seismic signal of an underground nuclear test explosion be “hidden” by the signal generated by a natural earthquake?

It’s possible, according to a new review article published in the Bulletin of the Seismological Society of America that contradicts the conventional wisdom about “masking.”

The new analysis by Joshua Carmichael and colleagues at Los Alamos National Laboratory found that advanced signal detector technology that can identify a 1.7-ton buried explosion with a 97% success rate only has a 37% when from that explosion are hidden within the seismic waveforms of an earthquake that happens within 100 seconds and about 250 kilometers away from the explosion.

“This is a very tiny object, with very weak gravity, so it easily loses a lot of mass, which then further weakens its gravity, so it loses even more mass,” said Dr. Avi Shporer.


What can a planet that’s shedding its material teach astronomers about planetary formation and evolution? This is what a recently submitted study to The Astrophysical Journal Letters hopes to address as an international team of scientists investigated a unique exoplanet that orbits its host star approximately 20 times closer than Mercury orbits our Sun, resulting in the exoplanet shedding so much material that it’s creating a tail of debris and will eventually disintegrate into nothing.

“The extent of the tail is gargantuan, stretching up to 9 million kilometers long, or roughly half of the planet’s entire orbit,” said Dr. Marc Hon, who is a postdoc in the Kavli Institute for Astrophysics and Space Research at the Massachusetts Institute of Technology (MIT) and lead author of the study.

Exoplanet BD+054868Ab is located approximately 140 light-years from Earth and orbits its star in approximately 30.5 hours. For context, Mercury takes our Sun in 88 days. The orbit of BD+054868Ab is so close, astronomers hypothesize that it’s a molten world slowly shedding its material and they estimate it will be completely gone between 1 million and 2 million years from now. During its long and slow death, BD+054868Ab is shedding so material that it’s leaving a trail of debris in its wake, which initially puzzled astronomers after analyzing data obtained from NASA’s Transiting Exoplanet Survey Satellite (TESS).