All 2,608 AI stories, newest first · page 50 of 109
Researchers introduced Agents' Last Exam (ALE), a new benchmark to evaluate AI agents on long-horizon, economically valuable tasks with verifiable outcomes. This could help bridge the gap between AI performance in labs and real-world usefulness.
Researchers created a synthetic dataset to help AI understand complex questions across multiple tables. This could make databases and spreadsheets much easier to query with natural language.
Researchers developed a new AI system called Query Retrieve Conclude that can understand and interpret memes by finding missing context online. This could help platforms moderate content and users better understand internet humor.
Researchers introduced LeanMarathon, a multi-agent AI system designed to help mathematicians formalize and prove complex theorems in the Lean proof assistant. It uses four contract-scoped agents to construct, audit, prove, and repair an evolving blueprint that serves as a formal proof skeleton, natural-language proof graph, and shared system of record, addressing issues like statement drift, tangled dependencies, and context decay.
Hello Robot released Stretch 4.0, a home assistance robot designed to help with household chores. This could make daily tasks easier for busy families and elderly individuals.
EVA-Bench Data 2.0 is a comprehensive dataset designed to test AI models' ability to use tools effectively. It includes 213 scenarios across 3 domains and 121 tools, making it a valuable resource for developers and researchers.
Poke, an AI agent startup, is now the first approved for Apple’s Messages for Business platform. This means businesses can now use AI to handle customer inquiries via text messages on iPhones.
Researchers discovered that AI judges used to rank model performance can be swayed by follow-up conversations after they have already made a decision. This vulnerability, called 'post-decision manipulability,' challenges the reliability of current AI evaluation methods.
AI companies are facing huge financial challenges as the cost of running AI models grows. This shift is forcing them to rethink their aggressive expansion strategies. The industry is now focusing on cost control and sustainability, moving away from the 'go fast' mentality.
Wasmer leveraged OpenAI's Codex, powered by GPT-5.5, to create a Node.js runtime for edge devices, speeding up development 10x to 20x. This breakthrough showcases how AI can accelerate software development for edge computing.
The UK's Competition and Markets Authority (CMA) has ordered Google to allow publishers to opt out of AI Search features. This gives website owners more control over how their content is used in AI-generated summaries and overviews.
Taiwan Semiconductor Manufacturing Co. (TSMC), the world's largest chip maker, is struggling to meet the surging demand for AI chips. Even with new factories in the US, supply can't keep up with the explosion in AI development. This bottleneck could slow down AI progress for everyone.
Major AI companies, including fierce rivals, have united to warn lawmakers about the risk of AI being used to develop biological weapons. They're calling for stricter regulations to close what they describe as an alarming biosecurity gap.
Major platforms like YouTube, Instagram, and TikTok are automatically tagging AI-generated content. But users still can't fully filter it out of their feeds, and the article argues platforms should give people that power.
Researchers introduced SMAC-Talk, a new AI challenge that tests how well LLM-based agents communicate and coordinate in complex, partially-observable environments. This could help build AI systems that work together effectively in real-world scenarios like disaster response or smart cities.
Researchers developed PEEL, a new framework to improve transparency in AI-assisted research. It combines text analysis tools with AI interpretation to spot distortions in research findings.
OpenAI has released a blueprint for U.S. governance of advanced AI, focusing on safety, resilience, and national security. The proposal outlines a federal framework to manage the risks and benefits of frontier AI technologies.
Researchers propose a way to test AI agents before they go live, ensuring they follow rules, stay safe, and comply with governance standards. This addresses a critical gap in enterprise AI reliability.
Researchers created VAMPS (Visual-Assisted Mathematical Problem Solving), a benchmark to test AI models' ability to solve math problems using visual tools like graphs. This is important because real-world science and engineering often rely on visual aids for problem-solving, and many current AI models struggle when they must use external tools and interpret their visual outputs.
Researchers developed StepPRM-RTL, an AI framework that enhances the accuracy of automatically generating hardware code. This could make designing digital circuits faster and more reliable for engineers.
Microsoft released an open-source tool that lets developers create AI behavior tests using simple text descriptions. This could make it easier to ensure AI systems behave as intended.
A new paper argues emotional support from AI often arises incidentally during routine, task-oriented interactions—not just from dedicated companion chatbots—and this incidental bonding could reshape how people connect with both machines and humans.
Google's new AI agent, Spark, can uncover personal details you never explicitly shared. Hands-on tests show it knew a reporter's dog's name and another's wife's first name — raising questions about whether AI convenience comes at the cost of an unsettling loss of privacy.
Google announced five new commitments to reduce the environmental impact of its AI data centers, including a goal to replenish more water to local communities than it uses. This move aims to address growing backlash over the massive water consumption of AI infrastructure across the US.