Publishing House Workers Report Increasing AI Use Amid Staff Resistance
Workers at three major publishers say AI is being used for cover art, copy, and emails, with some executives pushing junior staff to adopt the technology.
Sources · WIRED AI
The news desk
Follow what’s changing.
Recent developments in artificial intelligence, with the original sources one click away.
All news
101–120 of 1,892 Meta Releases AI Agent App After Months of Safety Delays, Citing Competitive Pressure
Meta launched its AI agent app Muse after extended safety reviews, with competition accelerating the decision, according to The New York Times.
Sources · NYT Technology
MIT Technology Review Questions Overreliance on AI Refusal Capabilities
MIT Technology Review examines concerns about excessive faith in AI systems' ability to decline harmful requests.
Sources · MIT Technology Review
OpenAI Releases 719 Math Proofs While Western Companies Launch Open-Weight Models
OpenAI published hundreds of math manuscripts from an unreleased model as Mistral and Reflection AI released new open-weight models.
Sources · Last Week in AI
Oracle Uses ChatGPT and Codex to Streamline Workflows Across Multiple Departments
Oracle has integrated ChatGPT Work and Codex to transform specialist workflows in recruiting, engineering, and operations, reducing task times significantly.
Sources · OpenAI
AI Robotics Breakthroughs Unlikely to Impact Daily Life Soon, Says MIT Technology Review
Despite recent visibility of humanoid robots, experts say AI robotics advances won't translate to real-world applications in the near term.
Sources · MIT Technology Review
Claude Haiku 5.5 Launches on Amazon Bedrock and AWS Claude Platform
Anthropic's Claude Haiku 5.5, designed for high-volume tasks, is now available on AWS with 75% cost reduction compared to its predecessor.
Sources · Amazon AWS AI
Researchers Develop Framework Enabling Frozen AI Models to Learn from Deployment Experience in Medical Applications
New model-agnostic framework allows frozen language and vision models to improve medical diagnosis performance by up to 34.2% through external expertise.
Sources · arxiv.org
Researchers Release Four New Benchmarks for Evaluating Large Language Models
New benchmarks test LLMs on Slovak language tasks, self-reporting accuracy, multimodal document reasoning, and credential security.
Sources · arxiv.org
Researchers Test AI Models Driving Real Toyota Corolla, Only One Succeeds
Three engineers tested GPT, Claude, and Grok controlling a real car for an In-N-Out run, with only one AI model successfully completing the task.
Sources · WIRED AI
Atlassian and OpenAI Expand Partnership to Connect AI Models with Enterprise Knowledge
Atlassian and OpenAI are deepening their collaboration to integrate frontier AI models with enterprise knowledge systems.
Sources · OpenAI
Mirror Particle Unveils World Model to Predict Human Behavior for Market Research
Mirror Particle is launching a world model designed to predict human behavior, arguing it outperforms LLM role-play for market research applications.
Sources · TechCrunch AI
Mistral Releases 1 Trillion-Parameter AI Model 'Le Chonk,' Claims Best Open-Weight Performance Outside China
French AI company Mistral released Mistral Large 4, a 1 trillion-parameter model it says competes with top US and Chinese AI systems.
Sources · wired.com / techcrunch.com
New Methods Tackle Data Quality and Training Challenges in Large Language Models
Recent arXiv papers introduce techniques for improving LLM training data curation, reinforcement learning approaches, and automated preprocessing pipelines.
Sources · arxiv.org
Researchers Propose Memory Canonicalization Framework to Address Cross-Model Interpretation Drift in LLM Memory Systems
New framework aims to standardize how different LLMs interpret stored memory objects, addressing emotional and factual consistency issues.
Sources · arxiv.org
Enterprise AI Agents Face Knowledge Gap Despite Data Abundance
AI agents struggle to understand organizational context beyond raw data, according to MIT Technology Review analysis.
Sources · MIT Technology Review
Four New ArXiv Papers Advance LLM Training Methods and Reasoning Analysis
Recent research tackles data curation for training, RL's impact on reasoning, automated preprocessing, and off-policy optimization challenges.
Sources · arxiv.org
Kevin Roose Confirms He Wrote His AI Book Without AI Assistance
The author of 'The AGI Chronicles' emphasized his book was written entirely by a human, not artificial intelligence.
Sources · WIRED AI
New Framework Addresses Memory Interpretation Drift Across Different LLM Models
Researchers propose memory canonicalization to address how different LLMs interpret identical stored memories with varying factual and emotional accuracy.
Sources · arxiv.org
OpenAI Outlines Approach to EU Text Watermarking Requirements
OpenAI details how it will implement text watermarking for EU regulatory compliance, starting with researcher access to detection tools.
Sources · OpenAI
