Lifeboat Foundation

Presented by VAST Data

With access to just a sliver of the 2.5 quintillion bytes of data created every day, AI produces what often seem like miracles that human intellect can’t match — identifying cancer on a medical scan, a viable embryo for IVF, new ways of tackling climate change and the opioid crisis and on and on. However, that’s not true intelligence; rather, these AI systems are just designed to link data points and report conclusions, to power increasingly disruptive automation across industries.

While generative AI is trending and GPT models have taken the world by storm with their astonishing capabilities to respond to human prompts, do they truly acquire the ability to perform reasoning tasks that humans find easy to execute? It’s important to understand that the current AI the world is working with has little understanding of the world it exists in, and is unable to build a mental model that goes beyond regurgitating information that is already known.

An innovative microscopy technique bridges the gap between field of view and resolution.

Hints from the bowhead whale genome published nearly a decade ago predicted that the mammals may use this alternate strategy (SN: 1/6/15). “But you need actual experiments to actually validate those predictions,” Tollis says.

In the lab, study coauthor Vera Gorbunova at the University of Rochester in New York and her colleagues ran an assortment of experiments on cells harvested from bowhead whale tissue, as well as on cells from humans, cows and mice.

The whale cells were both efficient and accurate at repairing double-strand breaks in DNA, damage that severs both strands of the DNA double helix. Whale repair restored broken DNA to like-new condition more often than cells from other mammals, the team found. In those animals, mends to the genome tended to be sloppier, like a poorly patched pair of jeans. The team also identified two proteins in bowhead whale cells, CIRBP and RPA2, that are part of the DNA repair crew.

Analysis of intracranial EEG from epilepsy patients suggests three neural patterns associated with spatial attention and consciousness, and implicate high-level visual areas and lateralized fronto-parietal networks in shaping human conscious experience.

For a metal meteorite to retain a magnetic field, its parent asteroid may need a cold rubble core to help drive an internal dynamo.

A beyond-standard-model interaction between neutrinos could show up in future supernovae observations.

The observation of self-heating in magnetically confined plasmas represents a milestone on the road to fusion reactors based on such plasmas.

A fusion reactor would generate electricity using the energy released by nuclear-fusion reactions occurring in a plasma. A key step in the race toward realizing the dream of such a reactor is the creation of a burning plasma—one in which the fusion reactions themselves supply most of the heating needed to keep the plasma at fusion-relevant temperatures. This step has recently been demonstrated for inertially confined plasmas [1, 2] (see Research News: Ignition First in a Fusion Reaction) but has so far remained elusive for magnetically confined ones. This goal could now be within reach thanks to direct evidence for fusion-induced heating of electrons in magnetically confined plasmas obtained by Vasily Kiptily and colleagues at the UK-based Joint European Torus (JET) facility [3].

The fusion of two heavy hydrogen isotopes—deuterium (D) and tritium (T)—presents the most promising path to a fusion reactor, both because of the relative ease in getting these isotopes to fuse and because of the large amount of energy released in each reaction. When D and T fuse, an alpha particle (a helium-4 nucleus) and a neutron are generated, carrying the released energy in the form of kinetic energy. The goal of achieving energy production from controlled fusion on Earth relies on the created alpha particles remaining in the plasma and heating the fusion fuel to keep the reactions going, while the kinetic energy of neutrons escaping the plasma is converted to electrical energy.

Programming errors on the blockchain can mean $100 million lost in the blink of an eye. Ronghui Gu and his company CertiK are trying to help.

Pigs are generally considered to have high intelligence, and new research shows that they may also be empathic to other members of their social groups, helping them during instances of need. But is this behavior truly unselfish, or is it driven by goal-specific motivations?

To investigate, researchers from the Research Institute for Farm Animal Biology in Germany and Austria’s University of Veterinary Medicine Institute of Animal Welfare Science have studied helping behavior among social groups of domestic pigs. Their work, titled “Spontaneous helping in pigs is mediated by helper’s social attention and distress signals of individuals in need,” is published in Proceedings of the Royal Society B.

In the wild, animals will often help members of their social groups in times of distress, defending them against predators or releasing them from traps, snares, or other types of confinement. There is no general consensus over whether such helping behavior is truly empathic, or whether it might be driven by a more selfish motivation.