New AI System Lets Robots Learn from Video Like Humans
Researchers at Rice University have developed a system that allows robots to learn complex tasks by watching videos. This could make robots more adaptable and useful in everyday environments.
#ai
AI
1854 stories tagged AI · page 39 of 78
New AI Security Flaw Lets Hackers Take Over Your Computer
Microsoft researchers discovered a vulnerability called AutoJack that lets attackers execute arbitrary code on the host system running certain AI agents when a user simply visits a malicious web page.
New AI Research Could Make Chatbots Ask Better Questions
Researchers propose a new way for AI to understand and communicate uncertainty, which could help chatbots ask better questions. This could make AI assistants more helpful in everyday situations where information is unclear.
Amazon MGM Drops Film About OpenAI CEO Sam Altman
Amazon MGM has abandoned a film about OpenAI CEO Sam Altman, starring Andrew Garfield. The movie, titled Artificial, was set to depict Altman's dramatic firing and reinstatement in 2023.
AI Models Help Researchers Sift Through Medical Studies Faster
Researchers used AI to automatically identify studies with health quality data in PubMed. This could make medical research reviews faster and more consistent.
AI Language Models Show No Special Advantage for Related Languages
Researchers found that large language models don't transfer knowledge better between related languages like Arabic and Hebrew, regardless of model size. This challenges assumptions about linguistic similarity aiding AI performance. The study tested models with 4 billion to 671 billion parameters and various architectures.
AI Data Centers Get Priority Grid Access, But Supply Issues Remain
The U.S. government has ordered faster grid connections for AI data centers, but this doesn't solve the underlying electricity shortages. Experts warn it could worsen shortages for other customers and drive up costs.
US Government Bans Anthropic's Fable 5 Over Security Fears, Industry and Experts Push Back
The US government blocked the release of Anthropic’s latest AI models, Fable 5 and Mythos 5, citing national security concerns. Cybersecurity experts have signed an open letter arguing the restrictions are misguided and dangerous, while Anthropic itself notes that similar vulnerabilities exist in other AI systems available on the market.
Researchers Unify Three Core AI Architectures into One Powerful System
A new paper introduces the Integral Transform Network (ITNet), showing that convolutional networks, recurrent networks, and transformers are all variations of a single mathematical concept: a learnable integral transform. This unification could simplify AI development and lead to more versatile models.
Pentagon Used Elon Musk's Grok AI to Fire 2,000 Missiles at Iran, Official Says
The Pentagon reportedly used Elon Musk's Grok AI to launch 2,000 missiles at Iran. This marks a significant milestone in the use of AI in military operations, raising questions about autonomy and control in warfare.
New Tool Reveals Hidden Biases in AI Language Models
Researchers have developed TreeTracer, a visual analytics tool that reveals hidden biases in AI language models by analyzing multiple possible outputs instead of just one. This approach uncovers representational and syntactic biases that standard auditing methods miss, making AI fairness assessment more thorough.
New Study Reveals Flaws in AI Judges' Reliability
A large-scale study found that AI judges often overstate their accuracy, relying on flawed metrics that don't correct for chance agreement. The research evaluated 21 judges across 118 runs and over 541,000 judgments, revealing significant issues with reliability and bias.
New Study Reveals AI's Blind Spots in Medical Data Analysis
Researchers found that large language models (LLMs) often overestimate their confidence when analyzing medical data. The study suggests new methods to help AI recognize what it doesn't know, improving reliability in healthcare applications.
New Research Reveals Hidden Influences in AI Group Discussions
Scientists have uncovered how AI agents influence each other during group discussions, revealing a hidden 'herd effect' that shapes group decisions. The study models how AI agents balance their own internal beliefs with the pull of the group, akin to human social dynamics. This discovery could improve how AI systems make decisions by better mimicking human behavior and could lead to more reliable multi-agent AI systems.
New Research Compares Diffusion Language Models to Traditional AI
Researchers have analyzed how Diffusion Language Models (DLMs) differ from traditional AI models. DLMs generate text by gradually refining entire sequences, potentially offering new advantages over current methods.
New Protocol Helps AI Systems Spot and Fix Bad Answers
Researchers developed the Argent Signaling Protocol (ASP) to help AI systems distinguish between incomplete and completely wrong answers. This could make AI assistants more reliable and trustworthy.
New AI Training-Free Pruning Method Preserves LLM Reasoning While Slashing Costs
Researchers have developed Causal Attribution Pruning (CAP), a training-free method that prunes large language models by identifying critical attention heads through their causal impact on reasoning tasks. This reduces inference costs without sacrificing multi-step reasoning performance.
New AI System Improves Medical Record Analysis at University Medicine Essen
Researchers developed an AI tool called ACIE that better handles complex medical records. It addresses key challenges in retrieving and understanding patient data across multiple documents. This could lead to more accurate and efficient medical diagnoses and treatments.
New AI Method Uses Eye Scans to Predict Alzheimer's Risk
Researchers developed REVEAL++, an AI method that analyzes retinal fundus images and structured clinical narratives to predict Alzheimer's disease risk. The approach improves upon prior work by introducing differentiable phenotypic grouping for more precise risk stratification.
New AI Benchmark Tests Machines' Ability to Resolve Ambiguous Sentences Using Visual Clues
Researchers created LaViSA, a benchmark designed to test whether AI models can resolve structurally ambiguous sentences by leveraging visual scenes. This helps machines better interpret complex, real-world situations where word order alone isn't enough.
Hugging Face Brings AI Models to Real Robots with Strands and LeRobot
Hugging Face has integrated AI models with robot hardware, allowing developers to deploy models directly to robots. This breakthrough makes advanced AI capabilities accessible to physical robots for the first time.
DeepSeek-V4: AI Models Handle 1 Million Tokens in Context with Innovative Architecture
DeepSeek has released a preview of its V4 models, which can process up to 1 million tokens in context and introduce a new hybrid attention architecture. This breakthrough could make AI assistants much more useful for long documents and complex tasks. The models use new techniques to handle large amounts of text efficiently, making them faster and more capable than previous versions.
Can AI Run a Simulated Startup for 500 Days? CEO-Bench Tests the Limits
Researchers created a simulated startup to see if AI could handle day-to-day business decisions. The experiment reveals both the potential and limitations of AI in management roles.
Beyond LoRA: New Open-Source Technique Challenges AI Fine-Tuning Leader
Researchers have developed new open-source methods that outperform LoRA, the most popular AI fine-tuning technique. These approaches could make customizing large language models more accessible and affordable for everyone.