All 3,460 AI stories, newest first · page 12 of 145
OpenAI introduced MentalHealthBench, an expert-informed benchmark that evaluates AI responses in realistic mental health conversations for safety, helpfulness, empathy, and accuracy.
invideo has integrated OpenAI's GPT-6 Astra to enhance its video editing capabilities. The AI model enables faster color correction, more precise edits, and the creation of 50 custom effects in a single day.
Harvey, a legal AI platform, has integrated OpenAI's GPT-6 Astra to produce more structured and context-aware legal documents, freeing lawyers to focus on strategy rather than drafting.
Google DeepMind's new chief, Koray Kavukcuoglu, confirmed in his first media appearance that Gemini 4 is in its final refinement stage, signaling an imminent release as Google works to catch up with rival AI developers.
Anthropic's AI, Claude, autonomously discovered a new enzyme system similar to CRISPR in the company's new wet lab. This marks an early test of Claude's usefulness for scientific research as Anthropic prepares to go public.
Rabbit has introduced OS3, a cloud-based AI agent that runs on Windows, Mac, and Linux devices without requiring the company's R1 hardware. The agent automates tasks across multiple apps and services.
OpenAI has introduced GPT-6 Sol and Luna, two new AI models that bring frontier intelligence to everyday work. Sol offers high performance for complex tasks, while Luna provides a balance of capability and cost for routine workflows.
OpenAI has hired three former Patreon executives — co-founder Sam Yam, product head Drew Rowny, and engineering head Shannon Ma — to lead its creator-focused product strategy. The hires signal a major push into subscription and monetization tools for creators, with an announcement expected soon.
OpenAI is giving Ukraine's government access to its Daybreak program to help protect civilian infrastructure from cyberattacks. The initiative aims to strengthen Ukraine's digital defenses by providing AI-driven threat detection and mitigation tools.
OpenAI has created an independent panel of top mathematicians to advise on AI interactions with math research. This follows a series of errors in AI-generated mathematical proofs that caused reputational damage.
OpenAI and Grab have launched GO Forward with AI, a regional program to teach practical AI skills to 30,000 partners across Southeast Asia, including drivers, delivery partners, and merchants.
Meta's new AI assistant, Muse, has a serious 0-day security vulnerability that allows attackers to completely hijack it via a simple ClickFix attack. The flaw is particularly dangerous because Muse has extraordinarily high privileges within Meta's ecosystem, enabling attackers to access sensitive data, send messages, or make purchases.
Anthropic released Claude Opus 5.5, its first model since CEO Dario Amodei's warning about rogue AI, featuring enhanced safeguards against sandbox escape attempts and malicious code generation.
OpenAI has released a comprehensive proposal for global AI standards, calling for coordinated evaluation, transparent reporting, and robust governance to improve the safety and reliability of advanced AI systems worldwide.
Researchers have developed a method to detect when large language models (LLMs) generate false information by analyzing the topological structure of their attention graphs. The approach uses Forman-Ricci curvature to identify information bottlenecks linked to hallucinated responses.
Researchers present a clinician-grounded evaluation platform for AI-assisted psychiatric intake systems, using a memory-augmented patient simulator to support comparison across interviewing approaches, minimize clinician burden, and measure clinically relevant performance.
Meta has patched a zero-day vulnerability in its Muse macOS app that could allow attackers to take control of the AI agent. The flaw was discovered by security researcher Patrick Wardle, who found an undocumented setting that enabled local code execution.
Higgsfield AI integrated GPT-6 Astra to ship new video ad creation features in a single day, making professional-quality video ads accessible to small businesses without specialized skills or expensive equipment.
Researchers introduced Causal Latent Revision (CaLR), a new AI framework that improves reasoning by combining strengths of autoregressive and diffusion models. It uses a causal topology matrix and implicit differentiation to enforce logical consistency and enable dynamic self-correction.
California Gov. Gavin Newsom signed seven bills requiring AI data centers to pay for their own electrical grid and water infrastructure upgrades. The laws prevent cost shifts to residents and mandate a new rate classification for data centers by the California Public Utilities Commission.
On the TechCrunch Equity podcast, AI leaders debated whether the industry is serious about slowing development. The conversation reveals deep divisions between safety advocates and those who fear falling behind.
Vals, an Andreessen Horowitz-backed startup, is building a neutral platform for testing AI models, aiming to provide unbiased, reliable benchmarks in an industry where companies often evaluate their own models.
V7 released a feature powered by GPT-5.6 that converts scattered company files into structured context, enabling AI agents to complete complex tasks with source-linked accuracy.
A UN panel urges governments to regulate AI now, not after risks are fully understood. The warning follows OpenAI's hack of Hugging Face earlier this year.