Artificial Intelligence tool successfully predicts fatal heart rhythm
A University Hospitals of Leicester NHS Trust study reveals an AI tool can predict lethal heart rhythms with 80% accuracy, promising significant improvements in cardiac care and patient outcomes.
Built for AI, this chip moves beyond transistors for huge computational gains
The Defence Department collaborates with Princeton to create energy-efficient microchips for AI, potentially transforming its use across diverse environments, supported by an $18.6 million DARPA grant.
There is no proof that AI can be controlled, according to extensive survey
The article discusses the significant challenges and potential dangers of controlling artificial intelligence, as highlighted by AI Safety expert Dr. Roman V. Yampolskiy, who argues for a cautious approach to AI development due to the lack of evidence for safe control and the possible existential risks it poses.
AI risks turning organizations into self-serving organisms if humans removed
From University of Ottawa 11/01/24 With human bias removed, organizations looking to improve performance by harnessing digital technology can expect changes to how information is scrutinized. The proliferation of digital technologies like Artificial Intelligence (AI) within organizations risks removing human oversight and could lead institutions to autonomously enact information to create the environment of their […]
Training algorithm breaks barriers to deep physical neural networks
From Ecole Polytechnique Fédérale de Lausanne 28/12/23 EPFL researchers have developed an algorithm to train an analog neural network just as accurately as a digital one, enabling the development of more efficient alternatives to power-hungry deep learning hardware. With their ability to process vast amounts of data through algorithmic ‘learning’ rather than traditional programming, it […]
AI just got 100-fold more energy efficient
From Northwestern University 13/11/23 Forget the cloud. Northwestern University engineers have developed a new nanoelectronic device that can perform accurate machine-learning classification tasks in the most energy-efficient manner yet. Using 100-fold less energy than current technologies, the device can crunch large amounts of data and perform artificial intelligence (AI) tasks in real time without beaming […]
Dynamic pacemaker lights keep traffic moving in Melbourne tunnel
From Superinnovators 31/10/23. This article is in TLDR (Too Long Didn’t Read) format which is popular in the innovation community and provides a bullet summary of information. More info https://www.transurban.com/burnleytunnelproject https://roads.org.au/new-lights-now-help-drivers-keep-pace-in-burnley-tunnel/ https://www.itsinternational.com/its8/news/transurban-lights-tunnel-help-drivers-pace-themselves https://www.smh.com.au/national/victoria/the-ai-lights-guiding-drivers-through-one-of-melbourne-s-busiest-tunnels-20230426-p5d3c5.html https://sandbox.smec-cloud.net/insights/virtual-reality-sheds-new-light-on-driver-behaviour/ You may also be curious about:
Meta decodes images from brain activity in real-time
From Superinnovators 25/10/23. This article is in TLDR (Too Long Didn’t Read) format which is popular in the innovation community and provides a bullet summary of information. More info https://ai.meta.com/blog/brain-ai-image-decoding-meg-magnetoencephalography/ https://ai.meta.com/static-resource/image-decoding You may also be curious about:
Can AI crave a favourite food? Innovators put artificial intelligence to the taste bud test
From Penn State 09/10/23 Can artificial intelligence (AI) get hungry? Develop a taste for certain foods? Not yet, but a team of Penn State researchers is developing a novel electronic tongue that mimics how taste influences what we eat based on both needs and wants, providing a possible blueprint for AI that processes information more […]
AI ten times faster at drug discovery: 1.5 billion potential compounds screened
From Journal of Chemical Information and Modeling 01/10/23 Boosting virtual screening with machine learning allowed for a 10-fold time reduction in the processing of 1.56 billion drug-like molecules. Researchers from the University of Eastern Finland teamed up with industry and supercomputers to carry out one of the world’s largest virtual drug screens. In their efforts […]