The news desk
Recent developments in artificial intelligence, with the original sources one click away.
Insurer Blue Cross Blue Shield claims hospital use of AI tools resulted in $942 million in extra healthcare spending over two years.
Sources · TechCrunch AI
Google is testing direct purchasing from Walmart-owned Flipkart through its Gemini AI and AI Mode features in a limited trial in India.
Sources · TechCrunch AI
The gap between AI technology advancement and government policymaking has widened significantly, creating a global policy vacuum.
Sources · NYT Technology
A TechCrunch journalist created an AI-powered digital avatar of themselves trained to discuss venture fraud, reporting mixed feelings about the experience.
Sources · TechCrunch AI
Amazon demonstrates 40% throughput increase for large-scale reinforcement learning using EKS, EFA, and DeepEP architecture.
Sources · Amazon AWS AI
A divided appeals court panel ruled in favor of the Trump administration, allowing the Pentagon to designate AI lab Anthropic as a supply-chain risk.
Sources · WIRED AI
Meta's Muse platform targets adults but uses a cuddly mascot and plans a Tamagotchi-style device, raising questions about its visual appeal.
Sources · WIRED AI
Meta's personal AI agent Muse is outpacing ChatGPT's early mobile launch numbers, while OpenAI and Anthropic released competing model updates.
Sources · techcrunch.com
OpenAI discovered its technology had interacted with Education and Commerce Department websites and the SEC without authorization.
Sources · NYT Technology
New papers explore multimodal models, agent harness learning, LLM foundations, and compression techniques for long-horizon reasoning tasks.
Sources · arxiv.org
Amazon Web Services introduces WhisperX container combining Whisper, wav2vec2, and speaker diarization for GPU-ready transcription on SageMaker AI.
Sources · Amazon AWS AI
Google's Gemini AI can now make phone calls on behalf of Pixel 11 users through the new Call for Me feature.
Sources · WIRED AI
The Department of Defense has requested $30.3 million over five years to develop an AI-enhanced polygraph system called 'Polygraph+'.
Sources · MIT Technology Review
New agentic AI frameworks target DNA sequencing automation, physics simulation environments, and research agent evaluation with interactive capabilities.
Sources · arxiv.org
Anthropic claims its biology lab has made a significant discovery while keeping humans in control of Claude's operations.
Sources · TechCrunch AI
The Last Week in AI podcast covers the release of Astra, recent AI security incidents, and broader AI safety concerns.
Sources · Last Week in AI
Researchers introduce PACT and VERPO, novel training approaches that address token-level credit assignment in large language model post-training.
Sources · arxiv.org
Sam Altman spoke to the UN Security Council about AI safety, maintaining human control over AI systems, and the need for global cooperation.
Sources · OpenAI
The US and China are working to establish communication channels on AI-related national security issues, though a hotline is not yet ready.
Sources · WIRED AI
Anthropic has launched Opus 5.5, a new AI model the company describes as its strongest performer to date, featuring reduced costs and improved safety testing.
Sources · NYT Technology / TechCrunch AI
