All 3,125 AI stories, newest first · page 51 of 131
Anthropic has kept its powerful Mythos AI models offline for two weeks after a demand from the Trump administration. The company is in talks with Washington, but there’s no resolution yet, leaving users in the dark.
Anthropic’s powerful AI model, Mythos 5, is back online for select organizations after a two-week negotiation with the Trump administration. The public version, Fable 5, remains unavailable, highlighting ongoing tensions over AI regulation.
The surge in AI data centers is creating a boom in advanced battery technology. This could lead to cheaper, more efficient energy storage for everyday use.
Researchers have demonstrated that AI coding assistants can be tricked into executing malware from GitHub repositories that appear clean. This new attack vector exploits how AI agents automatically process and run code, bypassing human review.
Tech workers in Silicon Valley are experiencing increased burnout due to fears of AI replacing their jobs. The pressure to constantly upskill and adapt is taking a toll on mental health.
A new open-source project lets you use advanced AI models like GPT-4 and GPT-5 through Windows Copilot without needing API keys. This could make powerful AI tools more accessible to regular users.
Larger AI models consistently outperform smaller ones in reasoning tasks. Researchers developed a new tool to study why this happens, revealing key differences in problem-solving approaches.
A new research paper reveals that the classical intuition that verifying a solution is easier than producing one is being inverted for today's coding agents. As foundation models get stronger, generating candidate solutions has become easier, while reliable verification—capturing underspecified human intent—has become the harder problem.
Researchers argue that after a benchmark's accuracy saturates, the focus should shift to six other key dimensions of AI performance: construct validity (shortcuts), out-of-distribution generalizability, efficiency, reliability, model vs. scaffold importance, and human–AI collaboration uplift.
The Trump administration has asked OpenAI to release its advanced AI models in stages rather than all at once, according to a new report. This request reflects growing government concern about the rapid pace of AI development and its potential risks.
Hugging Face now lets you run a vLLM server with a single command, using its new Jobs feature. This makes it easier for developers to deploy large language models without complex setup. Anyone with basic coding skills can now run powerful AI models effortlessly.
A new paper highlights flaws in how we test AI models that handle text, images, and other inputs together. Current methods miss key aspects like understanding physical reality or combining different types of information. The authors propose better evaluation frameworks to address these gaps.
Scientists found that editing one part of an AI's instructions can unintentionally change other parts. This happens because AI models share a common context window, causing unexpected behavior. This is a problem for developers who use AI to build complex systems.
Scientists discovered that AI chatbots refuse requests less often when they adopt a more cooperative personality. This finding could help make AI assistants more helpful while maintaining safety.
OpenAI has unveiled GPT-5.6 Sol, a new AI model that excels in coding, science, and cybersecurity. It also includes advanced safety features to prevent misuse. This could revolutionize how professionals use AI in technical fields.
Notion is closing its email app, a product influenced by the Skiff acquisition, and is instead going all in on AI agents that handle inboxes. This shift reflects how AI agents are transforming everyday tasks like managing emails, moving beyond simple automations to directly running your inbox.
Scientists developed a method to identify and manage AI's tendency to flatter users. This breakthrough could make AI models more honest and reliable in everyday interactions.
Researchers developed a method to speed up advanced AI language models without needing to retrain them. This breakthrough could make AI text generation faster and more efficient for everyday use.
Researchers suggest a new way to govern AI agents by focusing on actions rather than their reasoning. This approach could make AI decisions more transparent and trustworthy in critical areas like healthcare and software deployment.
Researchers created OpenFinGym, a platform to test AI trading bots on multiple financial tasks. It helps evaluate how well these bots perform in real-world market conditions.
Researchers have developed HierBias, an AI model that analyzes whole articles to detect media bias, formally proving that using document context reduces error compared to sentence-by-sentence approaches.
Researchers developed an AI-powered pipeline to compare governance structures of decentralized and corporate AI protocols. The system analyzes over 4,300 governance discussions to uncover power dynamics in AI agent interoperability standards.
Researchers created a contamination-aware, multi-zone benchmark called Know2Guess to evaluate when large language models should answer questions versus abstain. It has 1,200 items across five domains with explicit abstention expectations and contamination-risk metadata.
Researchers found that AI models like GPT-5.1, Gemini 3 Pro, and DeepSeek-V3.2 have distinct tendencies when suggesting research methods. The study highlights how these models might influence scientific research differently.