An Australian research team has developed a groundbreaking prototype of a quantum battery that defies conventional battery logic. This battery demonstrates the previously impossible characteristic of charging faster as its size increases. Traditional batteries are constrained by longer charging times with increased capacity, but the quantum battery has the potential to fundamentally alter this relationship.

The key to this technology lies in leveraging quantum physics phenomena, particularly quantum entanglement, to allow the battery's storage units to charge collectively rather than independently. This means energy is transferred simultaneously, leading to a charging time that decreases proportionally to 1/√N, where N is the system's size.

The current prototype uses a multilayer organic microcavity to trap light and is wirelessly charged by a laser. While the amount of energy it can store is still very low and its duration is only a few nanoseconds, making it far from practical, it is the first functional system capable of both charging and discharging. This advancement could pave the way for widespread applications in the future, such as energy systems for quantum computing, remote wireless charging, and even ultra-fast charging for electric vehicles.

👉 Read the full article on Gizmodo en Español

U.S. company Xairos Systems has achieved a breakthrough in secure quantum communication in free space. The company successfully established quantum and optical communication links over a distance of approximately 1.93 kilometers (1.2 miles) without relying on GPS or fiber optic cables. This marks a significant step in applying quantum communication, previously confined mostly to laboratories or fiber optics, to real-world outdoor environments.

This achievement was made possible by a compact quantum terminal called Ares. Ares integrates three functions that typically require separate systems: 10 Gbps optical data transmission, distribution of entangled photons for quantum communication, and high-precision synchronization based on Quantum Time Transfer (QTT) technology developed by Xairos. QTT technology enables atomic-level precise time synchronization and quantum encryption key distribution even when GPS or radio frequency signals are unavailable or jammed.

This technology is expected to have applications in defense, telecommunications, energy, and finance. Notably, it meets the requirements set by the U.S. Space Development Agency (SDA) for future communication and defense systems, potentially becoming a crucial alternative for protecting communication systems against cyberattacks or intentional signal jamming. Xairos aims to scale this system to greater distances in the future, eventually targeting global secure quantum communication via satellite links.

👉 Read the full article on Gizmodo en Español

An internal OpenAI model has reportedly 'disproved' a central conjecture of Paul Erdős's famous unit-distance problem, first posed in 1946. Will Sawin, a mathematician at Princeton University, reviewed and further refined the AI-generated proof, publishing his own paper on the improvements. This achievement suggests that AI could usher in a new era of mathematical advancements, with Fields medalist Tim Gowers calling it a 'milestone in AI mathematics.'

Erdős's conjecture stated that the number of pairs of points with unit distance in a set of N points in a plane would not grow faster than N to the power of 1. However, OpenAI's model, by utilizing algebraic number theory and algebraic number fields of growing degrees, discovered a construction that suggests this conjecture is false. This approach was previously untried by humans, indicating AI's potential to make 'logical decisions' beyond mere computation.

Sawin describes the AI's proof as an 'informal proof' that closely resembles those produced by mathematicians. He further refined the AI's argument, successfully deriving a specific value indicating that the number of unit distances grows faster than N to the power of 1.01. From this experience, Sawin views AI as a 'tool,' highly effective for literature search and proofreading, but still requiring human intervention for idea generation.

👉 Read the full article on Gizmodo