This demo uses a deep neural network and two generic cameras to perform 3D pose estimation. — matlab-deep-learning/pose-estimation-3D-with-stereo-camera
Get the latest international news and world events from around the world.
Raising New Minds: A Developmental Framework for AI Alignment
AI alignment may depend not only on how we control artificial intelligence, but on how we teach, socialize, and learn to live with the minds we create.
Laser pulses capture unexplored polaronic states
In an international experiment, researchers observed Jahn–Teller polarons—quasiparticles that could play an important role in future ultrafast spintronic devices. These polarons emerged within the crystal lattice of cobalt oxide that had been activated by carefully tailored laser pulses.
When a cobalt oxide crystal is exposed to carefully tailored laser pulses, they induce specific local distortions of the crystal lattice that strongly affect the material’s structural, electrical and magnetic properties. The correlative experimental approaches that revealed these unexpected properties of cobalt oxide were carried out by a large international team of scientists from the University of Pavia (Italy), the Swiss Federal Institute of Technology Lausanne, the Paul Scherrer Institute (Switzerland), the University of Texas at Austin, the Massachusetts Institute of Technology and Northeastern University (U.S.). The theoretical description of the phenomenon, which made it possible to uncover the nature of the observed oscillations, was developed by physicists from the Institute of Nuclear Physics of the Polish Academy of Sciences (IFJ PAN) in Cracow.
Chemical catalysts, battery electrodes, photovoltaic cells and semiconductor gas sensors—these are just some of the modern applications of cobalt oxide (Co₃O₄). Despite its simple chemical formula, the unit cell of its crystal lattice consists of 56 atoms: 24 cobalt and 32 oxygen. Depending on their position within the unit cell, the cobalt atoms exist here in two oxidation states.
Agentic AI bot helps scientists speak to robots, speeding up experiments
Researchers at the Department of Energy’s Pacific Northwest National Laboratory use a slew of autonomous robots to design and implement experiments. However, setting up an experiment on an autonomous lab robot is surprisingly slow. The effort requires a lengthy back-and-forth between a scientist and an engineer to design the experimental steps—a process that can take weeks.
To help researchers work more efficiently, a PNNL team developed a generative agentic AI that can quickly translate experimental goals into instructions for a laboratory robot. The translation agent, called AutoLabs, is currently designed to operate with Big Kahuna, an automated robot built by Unchained Labs that researchers use to study new and existing battery materials. The system can carry out multistep experimental workflows, including mixing, heating, stirring and filtering samples with minimal human intervention. By automating these processes, researchers can perform five to 10 times more experiments than would be practical by hand.
The team published a paper in Scientific Reports about AutoLabs, and the software is also available for other researchers to download on GitHub.
Inorganic nanoscale device behaves like a single neuron, opening doors for AI and retinal implants
McGill University researchers have developed a light-detecting nanoscale structure that mimics how a neuron processes information. The neuron-like behavior emerges from the materials themselves, reducing the energy demand associated with similar devices that rely on circuits or software.
Instead of capturing data first and processing it elsewhere, the device senses and interprets light in the same place, similar to how the eye processes visual information.
The researchers say the discovery could increase the efficiency of vision-based technologies like artificial retinas and smart optical sensors. It could also transform how artificial neural networks (ANNs), a foundation of machine learning, are built. The research is published in the journal Nanoscale.
Sleep deprivation increases levels of the synaptic density marker SV2A in the human brain
The synaptic homeostasis hypothesis posits that sleep is essential for restoring cerebral equilibrium by downscaling synaptic connections that progressively strengthen and accumulate metabolic costs during wakefulness. While previously supported only by preclinical animal models, a recent study provides direct in vivo evidence of this mechanism in humans. Researchers evaluated 40 volunteers, half of whom underwent 28 hours of continuous sleep deprivation, utilizing Positron Emission Tomography (PET) to quantify levels of the SV2A protein, a reliable biomarker for synaptic density. The findings revealed that prolonged wakefulness significantly elevated SV2A levels across multiple brain regions, most notably in the hippocampus and thalamus. Furthermore, during a subsequent two-hour recovery sleep period, these elevated SV2A levels were strongly correlated with enhanced slow-wave activity, a primary electrophysiological marker of deep sleep and homeostatic sleep pressure. These results validate the synaptic homeostasis hypothesis in humans, demonstrating a measurable biological link between sleep deprivation, the accumulation of neural connections, and the restorative drive for deep, slow-wave sleep.
The synaptic homeostasis hypothesis (SHY) [1– 4] posits that wakefulness promotes synaptic potentiation due to environmental interactions and learning [5]. The strengthening of connections during waking elevates energy consumption, results in the accumulation of proteins and receptors that compete for the limited anatomical space in the skull and diminishes the signal-to-noise ratios in the neuronal network, ultimately saturating the capacity for learning. Sleep allows for synaptic down-selection, preserving energy and network efficiency. While the SHY has been supported by anatomical and molecular studies in animals, human evidence has remained limited due to the invasive nature of most techniques for quantifying synaptic strength.
Studies in animals indicate that anatomical or molecular markers of synaptic strength increase during wake and decline during sleep [6]. Firing rates in rodents indicate increased cortical excitability during wakefulness and decreased cortical excitability during sleep. In humans, cortical excitability is an indirect measure of plasticity. Findings from studies using transcranial magnetic stimulation (TMS) translated the findings from the above-mentioned rodent studies (reviewed in [7]). However, some in-vitro and in-vivo studies of synaptic strength in animals reveal opposite results, which may be due to differences in the type of marker, examined brain regions, cortical layers, or housing of animals (reviewed in [8]).
Synaptic vesicle glycoprotein 2A (SV2A) [9] is an integral membrane protein located on synaptic vesicles. Recent advances in PET imaging with tracers such as [¹⁸F]SynVesT-1 enable the noninvasive measurement of SV2A binding in the living human brain [10,11], allowing new opportunities to examine state-dependent synaptic changes. However, whether this reflects presynaptic terminal number, vesicle complement, SV2A expression per vesicle, or excitatory/inhibitory-synapse composition cannot be resolved with in vivo imaging. While SV2A availability is commonly interpreted as a proxy measure of synaptic density, we refer to it here as ‘SV2A-indexed synaptic density’ to reflect this interpretation while acknowledging its underlying biological ambiguity.
Anthropic Co-Founder: “I’m Scared of the Technology I’m Building” | The Rest Is Politics
Jack Clark — co-founder of Anthropic, the company behind Claude — says something almost no AI executive will say out loud: \.
Ben Goertzel Just Revealed When AGI Will ACTUALLY Happen | Ep. 38
Ben Goertzel, the godfather of AGI research and CEO of SingularityNet, just dropped some mind-blowing insights about artificial general intelligence that will change how you think about AI forever. This isn’t your typical AI hype this is raw truth from someone who’s been building AGI for decades.
In this deep dive conversation, Ben reveals the shocking reality behind current AI limitations, why decentralized AI infrastructure is crucial for humanity’s future, and his honest timeline for when we’ll actually achieve AGI. Plus, he shares what it’s like running a global AI empire while living on a remote island accessible only by ferry.
Key Topics Covered:
The real timeline for AGI development.
Why current AI models aren’t actually intelligent.
How SingularityNet is building decentralized AI infrastructure.
The ASI Alliance and the future of artificial superintelligence.
Ben’s daily routine managing hundreds of AI researchers globally.
Why math and music drive breakthrough AI thinking.
⏰ Timestamps:
0:00 — Introduction to Ben Goertzel.
2:30 — Daily life of an AGI pioneer.
8:45 — Managing a global AI empire.
15:20 — The truth about current AI limitations.
25:10 — SingularityNet and decentralized AI
35:40 — When will AGI actually happen?
45:30 — The future of artificial superintelligence.
58:15 — Closing thoughts.
-️ Subscribe for more exclusive interviews with crypto and AI pioneers who are building the future. Hit the notification bell so you never miss our latest episodes!
💬 What do you think about Ben’s AGI timeline? Drop your thoughts in the comments below!
Is This the Key to Never Getting Old?
Awesome results and a new project to double mice lifespan. If I could fund one researcher right now it would be this man.
In this Conference talk, Dr. Greg Fahy presents stunning data from the TRIIM and TRIIM-X trials. His team has successfully regrown the human thymus in older adults, reversed epigenetic aging clocks by up to two years, and restored immune function to levels seen decades earlier.
Beyond the lab results, participants showed dramatic real-world improvements: 15% stronger muscles, 21% better VO2 max, and frailty scores dropping to near zero. Dr. Fahy also unveils the \.
Sugar-coated nanoparticles show promise for treating most aggressive form of brain cancer
Researchers at Oregon State University have potentially found a new way to treat the most aggressive form of brain cancer, glioblastoma, whose two-year survival rate is less than 30%.
The study, led by Oleh Taratula, Olena Taratula and Yoon Tae Goo of the OSU College of Pharmacy, addresses what they describe as the two most persistent obstacles to effective glioblastoma treatment: delivering therapeutic agents through the blood-brain barrier, the cell network that acts as a security checkpoint between the bloodstream and the central nervous system, and then getting those agents to preferentially target tumors.
In research published in the Journal of Controlled Release, the scientists demonstrate the novel treatment technique in a mouse model. They loaded lipid nanoparticles with genetic material that promotes tumor suppression, then coated the nanoparticles with a type of sugar. The result was a 50% median increase in glioblastoma survival time.