Artificial general intelligence is a theoretical form of AI that can reason as well as humans. AI researchers debate how close it is to reality.
A research team, led by Professor Hoon Eui Jeong from the Department of Mechanical Engineering at UNIST has introduced an innovative magnetic composite artificial muscle, showcasing an impressive ability to withstand loads comparable to those of automobiles. This material achieves a stiffness enhancement of more than 2,700 times compared to conventional systems. The study is published in Nature Communications.
Soft artificial muscles, which emulate the fluidity of human muscular motion, have emerged as vital technologies in various fields, including robotics, wearable devices, and biomedical applications. Their inherent flexibility allows for smoother operations; however, traditional materials typically exhibit limitations in rigidity, hindering their ability to lift substantial weights and maintain precise control due to unwanted vibrations.
To overcome these challenges, researchers have employed variable rigid materials that can transition between hard and soft states. Yet, the available range for stiffness modulation has remained constrained, along with inadequate mechanical performance.
A judge found that two media outlets alleging copyright violations hadn’t demonstrated that they’d been harmed by OpenAI removing “copyright management information” from its training data.
Francois Chollet, a prominent AI expert and creator of ARC-AGI, discusses intelligence, consciousness, and artificial intelligence.
Chollet explains that real intelligence isn’t about memorizing information or having lots of knowledge — it’s about being able to handle new situations effectively. This is why he believes current large language models (LLMs) have \.
In a new study in Physical Review Letters, scientists have demonstrated a method to control artificial microswimmers using electric fields and fluid flow. These microscopic droplets could pave the way for targeted drug delivery and microrobotics.
Researchers at Tampere University have created the world’s first soft touchpad capable of detecting the force, area, and location of contact without the need for electricity. This innovative device operates using pneumatic channels, making it suitable for environments like MRI machines and other settings where electronic devices are impractical. The technology could also be advantageous for applications in soft robotics and rehabilitation aids.
Researchers at Tampere University have developed the world’s first soft touchpad that is able to sense the force, area, and location of contact without electricity. That has traditionally required electronic sensors, but the newly developed touchpad does not need electricity as it uses pneumatic channels embedded in the device for detection.
Made entirely of soft silicone, the device contains 32 channels that adapt to touch, each only a few hundred micrometers wide. In addition to detecting the force, area, and location of touch, the device is precise enough to recognize handwritten letters on its surface and it can even distinguish multiple simultaneous touches.
Australia has served up a Secure Innovation Placemat [PDF].
The wide variance in the documents is by design: each Five Eyes nation chose its own approach, although the campaign is a coordinated effort that is billed as “consistent and consolidated advice reflecting both the globalized and interconnected tech startup ecosystem as well as the global nature of the security threats startups face.” And everybody uses placemats.
Whether this advice will break through the “move fast and break things” culture that many startups nurture is anyone’s guess. The Register has reported on security and resilience troubles in the early years at Uber and Lyft, GitLab, and at OpenAI.
The close relationship between AI and highly complicated scientific computing can be seen in the fact that both the 2024 Nobel Prizes in Physics and Chemistry were awarded to scientists for devising AI for their respective fields of study. KAIST researchers have now succeeded in dramatically shortening the calculation time of highly sophisticated quantum mechanical computer simulations by predicting atomic-level chemical bonding information distributed in 3D space using a novel approach to teach AI.
Chinese researchers with ties to China’s People’s Liberation Army (PLA) have built an AI model called ChatBIT, designed for military applications using Meta’s open-source Llama model. According to Reuters, some researchers are associated with the Academy of Military Science (AMS), the PLA’s top research group.
Three academic papers and several analysts have confirmed the information, with ChatBIT using Meta’s Llama 13B large language model (LLM). This LLM has been modified for intelligence gathering and processing, allowing military planners to use it for operational decision-making.
The discovery of the quantum tunneling (QT) effect—the transmission of particles through a high potential barrier—was one of the most impressive achievements of quantum mechanics made in the 1920s. Responding to the contemporary challenges, I introduce a deep neural network (DNN) architecture that processes information using the effect of QT. I demonstrate the ability of QT-DNN to recognize optical illusions like a human. Tasking QT-DNN to simulate human perception of the Necker cube and Rubin’s vase, I provide arguments in favor of the superiority of QT-based activation functions over the activation functions optimized for modern applications in machine vision, also showing that, at the fundamental level, QT-DNN is closely related to biology-inspired DNNs and models based on the principles of quantum information processing.
