Ethics & Regulation
The $28 Million Mistake That Inspired Estonia’s AI ‘Fuckup Finder’
Estonia developed an AI system called 'Fuckup Finder' to identify legal errors in government documents after a wording...
Wired AI
Healthcare AI
NHS AI blood test could reduce invasive womb cancer checks
NHS hospitals are implementing an AI-powered blood test to help assess women referred for potential womb cancer,...
AI News
AI AgentsOpenAI
ChatGPT is now a partner for your most ambitious work
OpenAI has launched ChatGPT Work, an AI agent capable of taking autonomous actions across applications and files to...
OpenAI Blog
Healthcare AIOpenAI
GPT-5.5 Bio Bug Bounty
OpenAI has launched a Bio Bounty program related to GPT-5.5, focusing on identifying and addressing potential...
OpenAI Blog
LLMOpenAI
GPT-5.6: Frontier intelligence that scales with your ambition
GPT-5.6 is announced as a frontier intelligence model offering improved efficiency with more capability per token and...
OpenAI Blog
Coding Co-pilotOpenAI
OpenAI's AI beats every human at AtCoder, a top competitive programming contest
OpenAI's AI system won an exhibition match at the AtCoder World Tour Finals 2026, solving all five competitive...
The Decoder
LLM
NVIDIA Releases Nemotron-Labs-3-Puzzle-75B-A9B: A Compressed Hybrid MoE LLM Delivering 2.03x Server Throughput at Matched User Throughput
NVIDIA released Nemotron-Labs-3-Puzzle-75B-A9B, a compressed hybrid MoE language model that reduces parameters from...
MarkTechPost
Coding Co-pilotGoogle
Google AI Studio Adds Import from GitHub to Build a Deployable App
Google AI Studio is introducing Import from GitHub functionality in Build mode, allowing developers to transform...
MarkTechPost
LLM
Datalab Lift vs the Field: How a 9B Schema-First Extractor Compares with NuExtract3, LlamaExtract, Marker, and Docling
Datalab's Lift is a 9B schema-first document extraction tool that converts PDFs and images directly to JSON without...
MarkTechPost
Research & Papers
Robbyant Releases LingBot-VLA 2.0: An Open-Source 6B Vision-Language-Action (VLA) Model for Cross-Embodiment Robot Manipulation
Robbyant has released LingBot-VLA 2.0, an open-source 6B vision-language-action model trained on 60,000 hours of robot...
MarkTechPost
LLMxAI
SpaceXAI Releases Grok 4.5, a Cursor-Trained Model for Coding, Agentic Tasks, and Knowledge Work at $2/M Input
SpaceXAI released Grok 4.5, a model trained on Cursor data for coding and agentic tasks, featuring 80 TPS throughput...
MarkTechPost
Startups & FundingMeta
Meta’s Muse Image Might be Just What SMBs Need
Meta has released Muse, a new AI imaging model that allows users to generate competitive ad content directly on Meta's...
AI Business
Startups & Funding
Messi and Ronaldo Are Building Tech Portfolios. Mo Salah Is Playing a Different Game
Lionel Messi and Cristiano Ronaldo are investing in AI and health tech startups as part of their diversified investment...
Wired AI
Ethics & RegulationGoogle
Google’s deepfake detector system used to debunk McConnell hoax pic
Google's deepfake detection system successfully identified and helped debunk an AI-generated fake image purporting to...
TechCrunch AI
LLM
I Built a Self-Improving AI, and So Can You
The article discusses experiments where AI is used to build and improve other AI systems, demonstrating that advanced...
Wired AI
AI Agents
Nvidia, Hugging Face Collaborate on Open Source Robot Models
Nvidia and Hugging Face have announced a collaboration to develop open source robot models, aimed at improving...
AI Business
LLMxAI
SpaceXAI releases Grok 4.5, which Elon describes as an ‘Opus-class model’
Elon Musk's xAI company released Grok 4.5, a new large language model positioned as an 'Opus-class' model. The release...
TechCrunch AI
Startups & Funding
This startup thinks robotics is about to have its ChatGPT moment
General Intuition is leveraging millions of hours of video game data to train foundation models for physical AI, aiming...
TechCrunch AI
LLMxAI
Grok 4.5 is so cheap compared to Fable 5 and GPT 5.5 that benchmark gaps may not matter much
xAI releases Grok 4.5, a large language model trained on tens of thousands of Nvidia GB300 GPUs that underperforms...
The Decoder