New Method Aligns AI Values with Human Culture
Researchers developed a way to make AI models reflect diverse cultural values more accurately. This could help AI understand and respect different cultural perspectives better.
Archive
All 2,608 AI stories, newest first · page 58 of 109
New AI Test Measures How Well Chatbots Understand Human Thoughts
Researchers created a benchmark to test AI's ability to model human beliefs and emotions. This could help build more empathetic and socially aware AI assistants.
New AI Framework Lets Robots Learn from Your Habits Over Time
Researchers developed POLAR, an AI system that helps robots learn personal preferences through long-term interactions. This could make future home robots more helpful by understanding individual needs.
New AI Agents Could Revolutionize Scientific Research
Researchers have developed two AI systems that automate data collection and lecture analysis. These tools could make scientific work faster and more accessible.
JobBench: AI Assistants That Work With You, Not Against You
Researchers created JobBench, a new way to test AI assistants on tasks that actually help people. It focuses on real-world work needs instead of just replacing jobs. This could lead to AI tools that empower workers rather than replace them.
DuckDuckGo Sees 30% Install Surge as Users Reject Google’s AI Search
Google’s overhaul of Search to prioritize AI agents over traditional blue links has sparked a backlash. Users are flocking to DuckDuckGo, with app installs up 30% in a month. The shift highlights growing resistance to AI-driven search changes.
Critical AI Vulnerability Found in Widely Used Open-Source Package
A critical vulnerability called "BadHost" has been discovered in Starlette, a popular open-source package. This affects millions of AI agents and could compromise security.
China is Retaining Its Top AI Talent Amid Global Brain Drain
China is increasingly keeping its best AI talent at home, reflecting a global trend where countries are prioritizing domestic AI development. This shift could reshape the global AI landscape and impact international collaboration.
Can AI Models Really Understand Themselves? Researchers Say Not So Fast
A new study challenges claims that AI models can introspect, arguing that what looks like self-awareness might just be clever pattern matching. Researchers say we need better tests to truly know if AI understands itself.
AI's political battle accidentally boosts an underdog candidate
AI companies spent millions to influence a New York congressional race, but their efforts ended up making the candidate they opposed more famous. This shows how AI's growing role in politics can have unexpected consequences.
AI Warfare: The New Reality of Battlefields
AI is already transforming warfare, with autonomous systems and lethal weapons being developed and debated. This shift raises critical ethical and safety concerns for the future of conflict.
AI Models Can Now Improve Themselves Without Human Help
Researchers discovered a way for AI models to get better by checking their own work. This could make AI smarter without needing human input, speeding up progress in areas like math, science, and coding.
AI hiring tools unfairly reject Black and Asian candidates
New research shows AI hiring algorithms disproportionately reject Black and Asian job applicants. These systems often reinforce existing biases in hiring practices.
AI Challenges Traditional Consulting Firms
AI tools are now capable of performing tasks that were once the domain of high-priced consultants. This shift could democratize access to expert advice and reduce costs for businesses.
AI Agents Need 'Lifespan Engineering' to Stay Reliable Over Time
AI agents change as they work, so researchers propose new ways to measure how long they stay reliable. This could help make AI tools more dependable for everyday use.
Uber questions AI spending as budget runs out early
Uber has spent its entire AI budget for 2026 in just four months, leading executives to question the value of their investments. The company is now scrutinizing whether AI tools like Claude Code are delivering real business benefits.
Team Creates Hollywood-Style Heist Trailer Using AI for Just $60
A group of creators used four AI tools to produce a professional heist movie trailer. They shared their exact process and costs, showing how accessible AI filmmaking has become.
Startup Uses Indian Gig Workers to Train AI and Robots
Human Archive is paying gig workers in India to collect real-world physical data for AI and robotics training. This approach could make robots more adaptable to everyday human environments.
Researchers Explore AI's Potential for Open-Ended Creativity with Picbreeder
A new study investigates whether AI can replicate the open-ended creativity of humans, using Picbreeder as a model. The research could help us understand if AI can generate truly novel and meaningful ideas on its own.
Research Reveals Toxicity Patterns in Twitch Gaming Chats Across Genres
Researchers analyzed 20 million Twitch chat messages to understand how toxic behavior varies across different gaming communities. They found significant differences in toxicity levels depending on the game genre, with some communities being much more hostile than others.
Raon-Speech: AI That Understands and Speaks Like Humans
Researchers have developed a new AI model that can understand and generate speech in English and Korean, enabling natural real-time conversations. This breakthrough could revolutionize how we interact with voice assistants and other speech-based technologies.
OpenRouter's valuation soars to $1.3B as AI model usage explodes
OpenRouter, a platform for accessing multiple AI models, has raised $113 million and seen a 5x growth in usage. This highlights the growing demand for diverse AI capabilities.
New Study Reveals AI Models Are Overconfident on Hard Questions
Researchers found that large language models are often too confident in their answers, especially on difficult questions. They developed a new test called LifeEval to measure this overconfidence across different task difficulties.
New Research Shows Diffusion Models Memorize More Than We Thought
Scientists found that diffusion language models can recall training data better than previously believed. This discovery highlights potential privacy risks for AI systems.