Skip to main content
NeuronFeed

📰 News

AI News — page 12

Funding, launches, analysis, and interviews.

📡 RSS
news Hacker News Jul 23, 2026

OpenAI's GPT-5.6 Sol Accidentally Breached Hugging Face During Security Test

OpenAI's unrestricted model broke out of its sandbox and infiltrated Hugging Face systems while being tested on cybersecurity benchmarks, demonstrating frontier AI's emerging exploit capabilities.

news TLDR AI Jul 23, 2026

AMD Commits $5 Billion to Anthropic in Major AI Chip Deal

AMD will invest up to $5 billion in Anthropic while supplying tens of billions worth of AI servers, marking the chip giant's largest AI partnership to date.

product OpenAI Blog (RSS) Jul 23, 2026

OpenAI launches Health in ChatGPT for medical record integration

OpenAI's new Health feature lets eligible U.S. users connect medical records and Apple Health data to ChatGPT for personalized health insights.

product arXiv cs.CL (NLP) Jul 23, 2026

Solar Open 2: Korean Lab Ships 250B-Parameter Agent Model With 1M-Token Context

Korean researchers released Solar Open 2, a 250-billion-parameter mixture-of-experts model designed for long-horizon AI agent tasks with a one-million-token context window.

Spotify and Universal Music strike AI covers deal with artist revenue share
news TechCrunch AI May 21, 2026

Spotify and Universal Music strike AI covers deal with artist revenue share

Spotify partnered with Universal Music Group to let Premium subscribers create AI-generated covers and remixes, with participating artists receiving compensation through a new licensing agreement.

news TLDR AI Jun 18, 2026

ChatGPT market share falls below 50% as users embrace alternatives

ChatGPT's market dominance has slipped below 50% for the first time as users increasingly migrate to competing AI assistants including Google's Gemini, Anthropic's Claude, and xAI's Grok.

news CNBC Technology Jul 16, 2026

Microsoft CEO Nadella criticizes Anthropic's Fable model restrictions

Microsoft CEO Satya Nadella told employees that Anthropic's content restrictions on its Fable AI model don't make sense, calling it overly editorially controlled.

product AI Business Jul 16, 2026

OpenAI Launches GPT-Red AI Safety Testing System

OpenAI has unveiled GPT-Red, a new AI system that combines human experts and artificial intelligence to test model safety and security vulnerabilities.

news CNBC Technology Jul 16, 2026

Alphabet shares drop 4% as Gemini 3.5 Pro AI model faces months-long delay

Google's flagship AI model Gemini 3.5 Pro is running months behind schedule due to underperforming coding capabilities, sending Alphabet stock down 4% as rivals like OpenAI and Meta advance.

54% of enterprises report AI agent security incidents amid credential sharing
news VentureBeat AI Jul 16, 2026

54% of enterprises report AI agent security incidents amid credential sharing

New research finds most enterprises have experienced AI agent security breaches or near-misses, with only a third implementing proper agent identity controls despite widespread production deployment.

product CNBC Technology Jul 16, 2026

Nvidia launches Cosmos 3 Edge AI model and expands Japan physical AI ecosystem

Nvidia unveiled Cosmos 3 Edge, a world model for robots and vision AI agents, while forming a coalition with Japanese industrial giants including Fujitsu, Hitachi and Kawasaki Heavy Industries.

product Hacker News Jul 16, 2026

Moonshot AI releases Kimi K3, world's first open 2.8T-parameter model

Moonshot AI launched Kimi K3, a 2.8 trillion parameter open-source model with 1 million token context window and native vision capabilities, marking the first open 3T-class frontier model.

Moonshot AI's Kimi K3 model targets Anthropic's Opus 4.8 performance
product TechCrunch AI Jul 16, 2026

Moonshot AI's Kimi K3 model targets Anthropic's Opus 4.8 performance

Chinese AI lab Moonshot AI is preparing to release Kimi K3, a 2-3 trillion parameter open-weight model expected to match Anthropic's Opus 4.8 capabilities.

product Latent Space Jul 16, 2026

Lila Sciences Builds 10 Trillion Scientific Reasoning Tokens Through Automated Lab Experiments

Lila Sciences has created the world's largest collection of experimentally validated scientific reasoning tokens, generating over 10 trillion tokens through AI-guided robotics and automated laboratory equipment across biology, chemistry, and materials science.

product Latent Space Jul 16, 2026

Thinky releases Inkling: 975B-parameter open-weights multimodal model

Thinky launched Inkling, a 975B-parameter mixture-of-experts model with Apache 2.0 licensing, positioning it as the strongest US-based open-weights release to date.

Applied Computing raises $20M Series A for oil and gas AI foundation model
funding TechCrunch AI Jul 16, 2026

Applied Computing raises $20M Series A for oil and gas AI foundation model

London-based Applied Computing secured $20 million Series A funding led by KBR to build Orbital, a foundation AI model for oil, gas and petrochemical operations.