A new study involving an AI scientist that can run its own experiments is both fascinating and terrifying, changing science forever.
Robotic automation has become a game-changer in addressing labor shortages. While traditional rigid grippers have effectively automated various routine tasks, boosting efficiency and productivity in industries that deal with objects of well-defined specifications, they fall short in sectors like the food industry, where delicate objects of varying sizes and shapes need to be handled. In these cases, a more specialized type of gripper is required.
“Bioinspired soft robotics seeks to develop technologies that draw inspiration from nature and leverage advanced materials and fabrication processes,” said Dr. Pablo Valdivia y Alvarado, Associate Professor at the Singapore University of Technology and Design (SUTD).
Soft grippers inspired by the natural dexterity and control of human hands are particularly well-suited to the food industry. They can adapt to objects of varying sizes and shapes while distributing forces more evenly, making them ideal for handling delicate items.
According to a paper published by Nature Computational Science on Friday, the researchers developed a model that bridges the gap between big, externally complex AI networks and the small, internally complex workings of the brain.
Industry experts said the team’s findings could mark a pivotal shift in AI development, prompting further exploration of computing solutions that are not dependent on silicon chips.
Designed to mimic human decision-making and physical interaction, the Astribot S1 robot can handle tasks that would traditionally require human dexterity and judgment.
Launched by Stardust Intelligence, a Chinese company, the robot has a human-like upper body structure mounted on a wheeled base.
Continue reading “China’s new humanoid robot does it all — from Kung Fu to cooking” »
A recent study by UC San Diego researchers brings fresh insight into the ever-evolving capabilities of AI. The authors looked at the degree to which several prominent AI models, GPT-4, GPT-3.5, and the classic ELIZA could convincingly mimic human conversation, an application of the so-called Turing test for identifying when a computer program has reached human-level intelligence.
The results were telling: In a five-minute text-based conversation, GPT-4 was mistakenly identified as human 54 percent of the time, contrasted with ELIZA’s 22 percent. These findings not only highlight the strides AI has made but also underscore the nuanced challenges of distinguishing human intelligence from algorithmic mimicry.
There’s a lot of chatter about AI supposedly passing the Turing Test. Don’t fall for it. Here’s the truth, including a close look at the Reverse Turing Test.
Scientists are rethinking how to implement automation for biologists to reduce costs, simplify adoption, and increase reproducibility.
Analog in-memory computing recent hardware implementations focused mainly on accelerating inference deployment. In this work, to improve the training process, the authors propose algorithms for supervised training of deep neural networks on analog in-memory AI accelerator hardware.
The brain-computer interface offers real-time feedback to boost rehab adherence.
Rehabilitation robots could help patients in the future by reading their neural activity via a headset.
Continue reading “Brainwave-reading robot ‘coaches’ could rehabilitate stroke survivors” »
Many fundamental processes of life, and their synthetic counterparts in nanotechnology, are based on the autonomous assembly of individual particles into complex patterns. LMU physicist Professor Erwin Frey, Chair of Statistical and Biological Physics at LMU Munich and member of the ORIGINS Excellence Cluster, investigates the fundamental principles of this self-organization.
