As millions turn to ChatGPT and other AI chatbots for therapy-style advice, new research from Brown University raises a serious red flag: even when instructed to act like trained therapists, these systems routinely break core ethical standards of mental health care. In side-by-side evaluations with peer counselors and licensed psychologists, researchers uncovered 15 distinct ethical risks — from mishandling crisis situations and reinforcing harmful beliefs to showing biased responses and offering “deceptive empathy” that mimics care without real understanding.
Neuromorphic computers modeled after the human brain can now solve the complex equations behind physics simulations — something once thought possible only with energy-hungry supercomputers. The breakthrough could lead to powerful, low-energy supercomputers while revealing new secrets about how our brains process information.
AI may learn better when it’s allowed to talk to itself. Researchers showed that internal “mumbling,” combined with short-term memory, helps AI adapt to new tasks, switch goals, and handle complex challenges more easily. This approach boosts learning efficiency while using far less training data. It could pave the way for more flexible, human-like AI systems.
A massive new study comparing more than 100,000 people with today’s most advanced AI systems delivers a surprising result: generative AI can now beat the average human on certain creativity tests. Models like GPT-4 showed strong performance on tasks designed to measure original thinking and idea generation, sometimes outperforming typical human responses. But there’s a clear ceiling. The most creative humans — especially the top 10% — still leave AI well behind, particularly on richer creative work like poetry and storytelling.
Scientists have discovered that the human brain understands spoken language in a way that closely resembles how advanced AI language models work. By tracking brain activity as people listened to a long podcast, researchers found that meaning unfolds step by step—much like the layered processing inside systems such as GPT-style models.
Foams were once thought to behave like glass, with bubbles frozen in place at the microscopic level. But new simulations reveal that foam bubbles are always shifting, even while the foam keeps its overall shape. Remarkably, this restless motion follows the same math used to train artificial intelligence. The finding hints that learning-like behavior may be a fundamental principle shared by materials, machines, and living cells.
Stanford researchers have developed an AI that can predict future disease risk using data from just one night of sleep. The system analyzes detailed physiological signals, looking for hidden patterns across the brain, heart, and breathing. It successfully forecast risks for conditions like cancer, dementia, and heart disease. The results suggest sleep contains early health warnings doctors have largely overlooked.
New research shows that AI doesn’t need endless training data to start acting more like a human brain. When researchers redesigned AI systems to better resemble biological brains, some models produced brain-like activity without any training at all. This challenges today’s data-hungry approach to AI development. The work suggests smarter design could dramatically speed up learning while slashing costs and energy use.
A new AI developed at Duke University can uncover simple, readable rules behind extremely complex systems. It studies how systems evolve over time and reduces thousands of variables into compact equations that still capture real behavior. The method works across physics, engineering, climate science, and biology. Researchers say it could help scientists understand systems where traditional equations are missing or too complicated to write down.
Spanish researchers have created a powerful new open-source tool that helps uncover the hidden genetic networks driving cancer. Called RNACOREX, the software can analyze thousands of molecular interactions at once, revealing how genes communicate inside tumors and how those signals relate to patient survival. Tested across 13 different cancer types using international data, the tool matches the predictive power of advanced AI systems—while offering something rare in modern analytics: clear, interpretable explanations that help scientists understand why tumors behave the way they do.
AI tools designed to diagnose cancer from tissue samples are quietly learning more than just disease patterns. New research shows these systems can infer patient demographics from pathology slides, leading to biased results for certain groups. The bias stems from how the models are trained and the data they see, not just from missing samples. Researchers also demonstrated a way to significantly reduce these disparities.
BISC is an ultra-thin neural implant that creates a high-bandwidth wireless link between the brain and computers. Its tiny single-chip design packs tens of thousands of electrodes and supports advanced AI models for decoding movement, perception, and intent. Initial clinical work shows it can be inserted through a small opening in the skull and remain stable while capturing detailed neural activity. The technology could reshape treatments for epilepsy, paralysis, and blindness.
Researchers combined deep learning with high-resolution physics to create the first Milky Way model that tracks over 100 billion stars individually. Their AI learned how gas behaves after supernovae, removing one of the biggest computational bottlenecks in galactic modeling. The result is a simulation hundreds of times faster than current methods.
Chimps may revise their beliefs in surprisingly human-like ways. Experiments showed they switched choices when presented with stronger clues, demonstrating flexible reasoning. Computational modeling confirmed these decisions weren’t just instinct. The findings could influence how we think about learning in both children and AI.
Researchers have created a prediction method that comes startlingly close to real-world results. It works by aiming for strong alignment with actual values rather than simply reducing mistakes. Tests on medical and health data showed it often outperforms classic approaches. The discovery could reshape how scientists make reliable forecasts.
USC researchers built artificial neurons that replicate real brain processes using ion-based diffusive memristors. These devices emulate how neurons use chemicals to transmit and process signals, offering massive energy and size advantages. The technology may enable brain-like, hardware-based learning systems. It could transform AI into something closer to natural intelligence.
Researchers at Tsinghua University developed the Optical Feature Extraction Engine (OFE2), an optical engine that processes data at 12.5 GHz using light rather than electricity. Its integrated diffraction and data preparation modules enable unprecedented speed and efficiency for AI tasks. Demonstrations in imaging and trading showed improved accuracy, lower latency, and reduced power demand. This innovation pushes optical computing toward real-world, high-performance AI.
A team at the University at Buffalo has made it possible to simulate complex quantum systems without needing a supercomputer. By expanding the truncated Wigner approximation, they’ve created an accessible, efficient way to model real-world quantum behavior. Their method translates dense equations into a ready-to-use format that runs on ordinary computers. It could transform how physicists explore quantum phenomena.
Scientists at Skoltech developed a new mathematical model of memory that explores how information is encoded and stored. Their analysis suggests that memory works best in a seven-dimensional conceptual space — equivalent to having seven senses. The finding implies that both humans and AI might benefit from broader sensory inputs to optimize learning and recall.
Scientists at Mount Sinai have created an artificial intelligence system that can predict how likely rare genetic mutations are to actually cause disease. By combining machine learning with millions of electronic health records and routine lab tests like cholesterol or kidney function, the system produces "ML penetrance" scores that place genetic risk on a spectrum rather than a simple yes/no. Some variants once thought dangerous showed little real-world impact, while others previously labeled uncertain revealed strong disease links.