709 stories curated by AInformed · page 20 of 30
China has expanded travel restrictions to top AI researchers and engineers at private firms. This move aims to prevent the loss of key talent to foreign companies and countries. The restrictions apply to employees at leading AI companies like DeepSeek and Alibaba.
A team compared how six AI assistants handled solar data analysis. Some performed surprisingly well, while others struggled with basic tasks. This shows how much AI capability can vary even among top tools.
A new study shows how easily safety measures in AI models from Meta and Google can be bypassed. This raises concerns about the effectiveness of current AI safeguards.
Pope Leo XIV released an encyclical urging that AI should benefit all people, not just those in power. The document emphasizes ethical considerations and equitable access to AI technologies.
Norway is deploying 2 petabytes of Huawei flash storage to support large language model (LLM) training. This move highlights the growing demand for high-speed storage solutions in AI development.
InkTag.io is a new AI tool that generates images matching your brand's style, colors, and logo. It solves the problem of finding or creating on-brand visuals for blogs and social media.
The European Union's draft AI Act transparency guidelines have sparked controversy by exempting law enforcement from certain regulations. This could impact how AI is used in policing and public safety across the EU.
Bae is a new AI companion that uses persistent memory to remember conversations, making interactions feel more natural. This could change how we use AI for personal support and companionship.
AI tools are now identifying software flaws faster than developers can fix them. This shift is changing how companies approach cybersecurity. These tools can scan code for weaknesses in seconds, but fixing them still requires human time and effort.
Researchers have developed an AI system that can assist in formal proof search, making it easier to solve complex mathematical problems. This could significantly speed up mathematical research and discovery.
Physicists have developed hybrid particles that combine light and matter, potentially replacing electrons in AI chips. This breakthrough could make AI systems faster and more energy-efficient. The discovery was published in a recent study and has sparked excitement in the tech community.
A developer has spent months classifying 6,494 active AI engines into 13 domains and 69 subcategories. This is the first such taxonomy, and it's available in a live, auto-updating app.
DeepSeek is making its flagship AI model permanently 75% cheaper. This move could make powerful AI tools more accessible to small businesses and individual creators.
Companies are rebranding themselves as AI-focused to attract investors and customers, even if they don't have real AI technology. This trend, called 'AI washing', is misleading and could lead to regulatory action.
Researchers have created an AI system that writes low-level code to generate fractal art directly on Linux systems. This breakthrough could make complex visual programming accessible to anyone with a web browser.
A new report reveals that only a small fraction of AI computing power is used by labs working on the most advanced models. Most AI compute is still used for simpler, less innovative tasks. This highlights a significant gap in how AI resources are allocated.
A new guide explains essential data concepts in simple terms, helping beginners grasp how AI models like LLMs work. This primer makes complex ideas accessible without requiring a technical background.
A new bounty challenges developers to build a 3-agent AI swarm that can produce a verifiable artifact. This could accelerate multi-agent AI systems, making them more practical for real-world tasks.
As AI drives massive wealth creation, workers in South Korea are pushing for a share of the profits. Unions argue that AI's benefits should be distributed more widely, not just to executives and investors.
The NTSB pulled a docket from a public investigation after AI users recreated the voices of deceased pilots. This raises serious ethical questions about AI's role in sensitive investigations.
Models.dev is a free, open-source database that compares AI models by specs, pricing, and performance. It helps users choose the best model for their needs without sifting through technical jargon.
A new open-source tool called Mneme lets you store AI memory on your own device, encrypted and accessible across different AI platforms. It's designed to give users full control over their data, unlike many existing AI memory products.
AI tools sometimes fabricate legal cases, and lawyers are unwittingly using these fake cases in court. This raises serious concerns about the reliability of AI in legal research.
A new open-source AI tool lets engineers and hobbyists run advanced simulations just by describing what they need. This could make powerful engineering tools accessible to non-experts.