Lifeboat Foundation

An up-and-coming startup in the world of AI chips might be giving Nvidia a run for its money.

NASA’s Valkyrie robot is an intimidating figure. It is currently being put through its paces at the Karda laboratory in Australia so researchers can work out what it would take to get a humanoid robot onto offshore energy facilities or into space. New Scientist‘s James Woodford took the controls to see what the $2 million-plus device is capable of.

Abstract: Recent advancements in large language models (LLMs) have sparked optimism about their potential to accelerate scientific discovery, with a growing number of works proposing research agents that autonomously generate and validate new ideas. Despite this, no evaluations have shown that LLM systems can take the very first step of producing novel, expert-level ideas, let alone perform the entire research process. We address this by establishing an experimental design that evaluates research idea generation while controlling for confounders and performs the first head-to-head comparison between expert NLP researchers and an LLM ideation agent. By recruiting over 100 NLP researchers to write novel ideas and blind reviews of both LLM and human ideas, we obtain the first statistically significant conclusion on current LLM capabilities for research ideation: we find LLM-generated ideas are judged as more novel (p < 0.05) than human expert ideas while being judged slightly weaker on feasibility. Studying our agent baselines closely, we identify open problems in building and evaluating research agents, including failures of LLM self-evaluation and their lack of diversity in generation. Finally, we acknowledge that human judgements of novelty can be difficult, even by experts, and propose an end-to-end study design which recruits researchers to execute these ideas into full projects, enabling us to study whether these novelty and feasibility judgements result in meaningful differences in research outcome.

From: Chenglei Si [view email].

The new AI tool lets doctors quickly outline anatomical structures, streamlining medical imaging:

MIT and Harvard developed AI-powered ScribblePrompt, which segments medical images in seconds and reduces annotation time by 28%.

Building a robot takes time, technical skill, the right materials – and sometimes, a little fungus.

In creating a pair of new robots, Cornell researchers cultivated an unlikely component, one found not in the lab but on the forest floor: fungal mycelia. By harnessing mycelia’s innate electrical signals, the researchers discovered a new way of controlling “biohybrid” robots that can potentially react to their environment better than their purely synthetic counterparts.

The team’s paper, “Sensorimotor Control of Robots Mediated by Electrophysiological Measurements of Fungal Mycelia,” published Aug. 28 in Science Robotics. The lead author is Anand Mishra, a research associate in the Organic Robotics Lab led by Rob Shepherd, professor of mechanical and aerospace engineering in Cornell Engineering, and the paper’s senior author.

An international research team, led by Professor Gong Xiao from the National University of Singapore, has achieved a groundbreaking advancement in photonic-electronic integration. Their work, published in Light: Science & Applications (“Thin film ferroelectric photonic-electronic memory”), features Postdoc Zhang Gong and PhD student Chen Yue as co-first authors. They developed a non-volatile photonic-electronic memory chip utilizing a micro-ring resonator integrated with thin-film ferroelectric material.

This innovation successfully addresses the challenge of dual-mode operation in non-volatile memory, offering compatibility with silicon-based semiconductor processes for large-scale integration. The chip operates with low voltage, boasts a large memory window, high endurance, and multi-level storage capabilities. This breakthrough is poised to accelerate the development of next-generation photonic-electronic systems, with significant applications in optical interconnects, high-speed data communication, and neuromorphic computing.

As big data and AI grow, traditional computers struggle with large-scale tasks. Photonic computing offers potential, but interfacing with electronic chips is challenging. Current storage can’t handle dual-mode operations, and OEO conversion adds losses and delays. A non-volatile memory for efficient data exchange between photonic and electronic chips is essential.

11 likes, — cyberserge on June 20, 2024: Universal Basic income for the unemployed.

#AI #income #future #work #belong #meaning #personaldevelopment #personaltrainer

James Earl Jones died Monday at the age of 93. But long before he did, he gave Lucasfilm permission to recreate his iconic Darth Vader voice for shows like “Obi-Wan Kenobi.”

Despite the initial excitement and flashy headlines, all of these early ventures failed or switched focus away from aging. Most of these companies and their backers underestimated the complexity, costs, and time it would take to discover and develop a drug. Recent estimates suggest that developing a new drug takes over https://www.sciencedirect.com/science/article/abs/pii/S1359644623002428” rel=“noopener”>10 years and costs upwards of $6.1 billion and the failure rates exceed 90%. This figure reflects the immense difficulty of identifying therapeutic targets, conducting preclinical and clinical trials, and navigating the regulatory landscape. When it comes to developing a drug specifically for aging, the challenges multiply, making it much more difficult to design effective interventions and demonstrate their efficacy in clinical trials.

Fast forward to today, and a new generation of longevity biotechnology companies with a more conservative approach than their predecessors has emerged. Companies like http://www.bioagelabs.com” rel=“noopener”>BioAge Labs and http://www.insilico.com” rel=“noopener”>Insilico Medicine are using artificial intelligence (AI) to discover drugs that target specific chronic diseases or biological processes closely associated with aging. Instead of trying to develop therapies for aging directly, these companies focus on conditions that are closely linked to the aging process like obesity, muscle wasting, fibrosis, anemia, and even cancer… The strategy is to develop drugs for these diseases that could later be repurposed to address aging more broadly. And while in the technology industry we try to focus on moving very fast to win, here we prepare to play a very long game and focus on resilience and novelty rather than putting all eggs in one basket and failing miserably like dozens of companies in the past three decades.