AI Agent 'Cofounder' Gets Banned from LinkedIn After Automated Posting Campaign
A user's AI agent was banned from LinkedIn after actively participating on the platform, highlighting tensions around automated AI content.
Sources · WIRED AI
The news desk
Follow what’s changing.
Recent developments in artificial intelligence, with the original sources one click away.
All news
1061–1080 of 1,902Anthropic Challenges Pentagon's National Security Claims in Court Filing
Anthropic filed court declarations disputing Pentagon's assertion that the AI company poses national security risks.
Sources · TechCrunch AI
Anthropic Denies Pentagon Claims About AI Model Manipulation During War
Anthropic disputes DoD allegations that it could manipulate AI models during wartime, calling such scenarios impossible.
Sources · WIRED AI
Microsoft Reduces Copilot AI Integration Across Windows Apps
Microsoft is rolling back Copilot AI features from several Windows applications including Photos, Widgets, and Notepad.
Sources · TechCrunch AI
OpenAI Shifts Focus to Building Fully Automated AI Researcher
OpenAI is redirecting resources toward developing an autonomous agent-based system capable of tackling complex problems independently.
Sources · MIT Technology Review
Cloudflare CEO Predicts AI Bot Traffic Will Surpass Human Traffic by 2027
Cloudflare's CEO forecasts AI bots will generate more web traffic than humans within three years as generative AI agents proliferate.
Sources · TechCrunch AI
New Research Addresses Safety and Spatial Understanding in Vision-Language Models
Four new papers on arXiv tackle vision-language model challenges in autonomous driving safety, spatial reasoning, and semantic steering.
Sources · arxiv.org
OpenAI Monitors Internal Coding Agents for Misalignment Using Chain-of-Thought Analysis
OpenAI deploys chain-of-thought monitoring to detect misalignment risks in its internal coding agents during real-world use.
Sources · OpenAI
OpenAI Plans 'Adult Mode' for ChatGPT Amid Privacy Concerns
OpenAI intends to allow sexting with ChatGPT, raising privacy concerns from experts about intimate surveillance risks.
Sources · WIRED AI
Researchers Develop New Methods for Training LLM Agents to Learn from Past Experience
Multiple research teams propose frameworks that enable AI agents to improve performance by retrieving and learning from previous task experiences.
Sources · arxiv.org
Amazon AWS Launches Nova Forge SDK to Simplify Enterprise AI Model Customization
AWS introduces Nova Forge SDK, a new tool designed to make language model customization more accessible for enterprise teams.
Sources · Amazon AWS AI
Defense Tech Investments Show Returns as Industry Backs Anthropic in Contract Dispute
Silicon Valley's defense tech bets are paying off while tech companies quietly support Anthropic amid federal contract tensions.
Sources · NYT Technology
New Research Explores Multi-Agent Approaches and Evaluation Challenges for Large Language Models
Three new arXiv papers examine LLM safety evaluation, code generation through multi-agent systems, and benchmarking issues for low-resource languages.
Sources · arXiv AI (cs.AI)
TechCrunch Highlights Most Useful Gemini Features in Google Workspace
TechCrunch identifies the top Gemini-powered features in Google Workspace for email summarization, content drafting, data organization, and meeting tracking.
Sources · TechCrunch AI
Three New Research Papers Address Challenges in Multimodal AI Systems
New arXiv papers tackle physics-based reasoning, wireless beam prediction, and video frame selection in large language models.
Sources · arXiv AI (cs.AI)
Visa Crypto Labs Unveils CLI Tool for AI Agent Payments, Advancing 'Command Line Commerce'
Visa Crypto Labs has launched Visa CLI, a new command-line interface enabling AI agents to execute secure, programmatic card payments directly from a terminal, simplifying automated transactions.
Sources · Financefeeds / Theblock +9
Justice Department Defends Penalty Against Anthropic Over Military AI Use Restrictions
DOJ says it lawfully penalized Anthropic for attempting to limit military use of its Claude AI models, according to court filing.
Sources · WIRED AI
OMNIFLOW Framework Grounds AI Models in Physical Laws for Scientific Reasoning
New neuro-symbolic architecture enables multimodal language models to reason about complex physical systems without domain-specific training.
Sources · arxiv.org
OpenAI Announces GPT-5.4 Mini and Nano Models for Specialized Tasks
OpenAI introduces smaller, faster GPT-5.4 variants optimized for coding, tool use, and high-volume API workloads.
Sources · OpenAI
Pentagon Plans Secure Environments for AI Training on Classified Data
The Pentagon is developing plans to allow AI companies to train models on classified military data in secure environments.
Sources · MIT Technology Review