<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel>
<title>ScoutHub Lab — AI news</title><link>https://scouthub.dev/en/news/</link><description>The most important global AI news each day, summed up in plain language with a link to the source, plus the newest open-source AI repositories.</description><language>en</language><lastBuildDate>Tue, 10 Nov 2026 01:27:12 GMT</lastBuildDate>
<atom:link href="https://scouthub.dev/en/news/feed.xml" rel="self" type="application/rss+xml"/>
<item><title>Anthropic proceeds with IPO plans and sets valuation range</title><link>https://techcrunch.com/2026/11/09/anthropic-reportedly-plans-to-move-forward-with-ipo-plans/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n867</guid><pubDate>Tue, 10 Nov 2026 01:27:12 GMT</pubDate><description>Anthropic plans to establish its valuation range this week and begin trading late next week. Some institutional investors are pushing for a valuation below the company&apos;s recent $1.5 trillion target. Market participants will watch the public debut closely to evaluate the ongoing strength of the AI sector. — TechCrunch</description><category>Business</category><source url="https://techcrunch.com/category/artificial-intelligence/">TechCrunch</source></item>
<item><title>Apple introduces $50 add-on packs for extra AI usage in Logic Pro</title><link>https://www.theverge.com/tech/1030084/apple-intelligence-usage-packs-ai-features-creator-studio-logic-pro</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n836</guid><pubDate>Mon, 09 Nov 2026 19:09:58 GMT</pubDate><description>Apple has added a $49.99 one-time purchase option in Logic Pro to provide additional access to AI capabilities. The pack allows creators to continue using Creator Studio AI features and the built-in assistant after hitting usage limits. This makes Logic Pro the company&apos;s first application to charge for extended Apple Intelligence usage. — The Verge</description><category>Business</category><source url="https://www.theverge.com/ai-artificial-intelligence">The Verge</source></item>
<item><title>ChatGPT Enterprise Is Now Available on AWS Marketplace</title><link>https://openai.com/index/chatgpt-enterprise-aws-marketplace</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n851</guid><pubDate>Mon, 09 Nov 2026 19:00:00 GMT</pubDate><description>ChatGPT Enterprise is now accessible directly through AWS Marketplace. This allows organizations to streamline purchasing by using their existing AWS billing accounts. — OpenAI</description><category>Business</category><source url="https://openai.com">OpenAI</source></item>
<item><title>Meta launches Muse Ember, a lightweight model for agent execution</title><link>https://techcrunch.com/2026/11/09/meta-releases-muse-ember-a-new-lightweight-model-for-agent-execution/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n874</guid><pubDate>Mon, 09 Nov 2026 18:00:00 GMT</pubDate><description>Meta has introduced Muse Ember, a lightweight and cost-efficient AI model tailored for executing repetitive agent tasks like tool calls, routing, and document parsing. Priced at 10 cents per million input tokens, the model handles text, audio, and visual data and is designed to work in tandem with larger models like Muse Wonder. — TechCrunch</description><category>Models &amp; releases</category><source url="https://techcrunch.com/category/artificial-intelligence/">TechCrunch</source></item>
<item><title>Claude Code Settings Files Leak Passwords and API Keys on GitHub</title><link>https://arstechnica.com/ai/2026/11/claude-code-settings-files-expose-thousands-of-secrets-on-github/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n924</guid><pubDate>Mon, 09 Nov 2026 17:24:26 GMT</pubDate><description>Security researchers at Apiiro found that Claude Code has exposed thousands of credentials, including API keys and passwords, in public GitHub repositories. The problem occurs when users approve shell commands to run automatically, storing sensitive commands inside local configuration files. Because these files are not excluded at the repository level, they can easily be committed and shared online by accident. — Ars Technica</description><category>Tools &amp; products</category><source url="https://arstechnica.com/ai/">Ars Technica</source></item>
<item><title>OpenAI Introduces Control Room to Monitor and Halt Research Model Workloads</title><link>https://openai.com/index/control-room</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n852</guid><pubDate>Mon, 09 Nov 2026 13:00:00 GMT</pubDate><description>OpenAI uses a central system called Control Room to track model workloads across its research infrastructure. The system is designed to rapidly stop these workloads if safety or security concerns emerge. — OpenAI</description><category>Tools &amp; products</category><source url="https://openai.com">OpenAI</source></item>
<item><title>New System Helps AI Agents Retain Behavioral Corrections Across Sessions</title><link>https://arxiv.org/abs/2611.08644</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n778</guid><pubDate>Mon, 09 Nov 2026 05:00:00 GMT</pubDate><description>Large language model agents typically fail to learn from corrections across separate sessions, repeating the same errors with users. To address this issue, researcher Nima Vahdat and a co-author developed Forge Decision Context (forge_dc). The system captures procedural knowledge and behavioral adjustments rather than focusing solely on basic facts and chat history. — arXiv cs.AI</description><category>Research</category><source url="https://arxiv.org/list/cs.AI/recent">arXiv cs.AI</source></item>
<item><title>AlterNet Architecture Dynamically Adjusts Compute Demands for LLMs</title><link>https://arxiv.org/abs/2611.08741</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n785</guid><pubDate>Mon, 09 Nov 2026 05:00:00 GMT</pubDate><description>Researcher Alexander Kastius proposed AlterNet, a new architecture combining a Mixture-of-Experts design with a dynamic-depth system. Unlike models with fixed compute requirements, this approach allows the system to adjust its forward pass and automatically determine the necessary computation. — arXiv cs.AI</description><category>Models &amp; releases</category><source url="https://arxiv.org/list/cs.AI/recent">arXiv cs.AI</source></item>
<item><title>Using Collapsible Tags to Manage Overly Detailed AI Pull Request Descriptions</title><link>https://simonwillison.net/2026/Nov/9/details-summary/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n808</guid><pubDate>Mon, 09 Nov 2026 04:25:05 GMT</pubDate><description>AI coding assistants like Claude Code and Codex generate accurate but excessively long pull request descriptions that can discourage developers from reading them. Simon Willison proposes writing a human-crafted summary and nesting the comprehensive AI-generated text inside collapsible HTML tags. Because GitHub Markdown supports these tags, developers can keep PRs clean while leaving full technical details available on demand. — Simon Willison</description><category>Tools &amp; products</category><source url="https://simonwillison.net">Simon Willison</source></item>
<item><title>Anthropic Introduces Claude Fable 5.2 and Claude Mythos 5.2 for Advanced Technical Work</title><link>https://www.anthropic.com/claude-fable-and-mythos-5-2</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-10/#n859</guid><pubDate>Mon, 09 Nov 2026 00:00:00 GMT</pubDate><description>Anthropic has released Claude Fable 5.2 and Claude Mythos 5.2, targeting complex technical work in areas like programming, finance, and scientific research. While both models share the same architecture, Fable includes broad safety controls for public use, whereas Mythos is restricted to approved organizations in cybersecurity and life sciences. The models generate responses up to 30% faster than prior versions and are priced at $8 per million input tokens and $40 per million output tokens. — Anthropic</description><category>Models &amp; releases</category><source url="https://www.anthropic.com/news">Anthropic</source></item>
<item><title>Apple accelerates Siri development to build a more capable AI agent</title><link>https://www.theverge.com/ai-artificial-intelligence/1029552/apple-siri-ai-agentic</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n762</guid><pubDate>Sun, 08 Nov 2026 19:18:19 GMT</pubDate><description>Apple leadership is pushing to speed up the development roadmap for Siri to better compete with rivals like OpenAI and Meta. Future versions of the assistant are expected to handle more complex tasks and control third-party applications. The effort builds on a functional version of Siri AI that was delivered on a very short timeline. — The Verge</description><category>Models &amp; releases</category><source url="https://www.theverge.com/ai-artificial-intelligence">The Verge</source></item>
<item><title>Elvis Presley and Eminem Rights Holders File Lawsuits Against Suno</title><link>https://www.theverge.com/ai-artificial-intelligence/1029533/elvis-eminem-suno-ai-music-clone-lawsuit</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n763</guid><pubDate>Sun, 08 Nov 2026 17:23:19 GMT</pubDate><description>AI music platform Suno is facing two new lawsuits from Elvis Presley&apos;s rights holders and Eminem&apos;s publisher. The legal actions accuse the company of generating unauthorized music and lyrics that illegally mimic both artists. Attorneys involved in the Presley case also claim that Suno violated laws in the state of Tennessee. — The Verge</description><category>Policy &amp; safety</category><source url="https://www.theverge.com/ai-artificial-intelligence">The Verge</source></item>
<item><title>How World Models Use Real-World Physics to Train Robots and Autonomous Cars</title><link>https://www.wired.com/story/world-models-will-soon-be-inescapable-what-you-need-to-know/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-09/#n804</guid><pubDate>Sun, 08 Nov 2026 10:00:00 GMT</pubDate><description>World models are shifting artificial intelligence from text processing into three-dimensional spaces. By simulating real-world physics, they help robots and autonomous vehicles learn to navigate reality. — WIRED</description><category>Research</category><source url="https://www.wired.com/tag/artificial-intelligence/">WIRED</source></item>
<item><title>Microsoft Research Reorganizes Its AI Frontiers Group into Microsoft AI</title><link>https://hunch.net/?p=13764006</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n78</guid><pubDate>Sat, 07 Nov 2026 18:02:17 GMT</pubDate><description>The team developed tools such as the AutoGen framework and the Fara computer agent while sharing code across the organization. In early September, the group was reorganized and integrated into the broader Microsoft AI division. — Hacker News</description><category>Business</category><source url="https://news.ycombinator.com">Hacker News</source></item>
<item><title>Concerns Raised Over US Approval of Nvidia AI Chip Sales to Chinese Firms</title><link>https://www.theverge.com/ai-artificial-intelligence/1029483/nvidia-ai-chips-china-blacklist-government</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n38</guid><pubDate>Sat, 07 Nov 2026 15:43:45 GMT</pubDate><description>Concerns have emerged that the US Commerce Department and Nvidia are not properly vetting Chinese firms cleared to buy H200 AI chips. One approved firm, New Matrix, was found to have completely empty offices. Additionally, authorities have approved sales to Chinese entities with ties to blacklisted companies. — The Verge</description><category>Policy &amp; safety</category><source url="https://www.theverge.com/ai-artificial-intelligence">The Verge</source></item>
<item><title>Open-Weight AI Models Narrow the Cyber Capability Gap with Proprietary Systems</title><link>https://the-decoder.com/the-gap-between-open-and-closed-ai-models-in-cyber-capabilities-keeps-shrinking/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n90</guid><pubDate>Sat, 07 Nov 2026 15:35:50 GMT</pubDate><description>Since open models cannot be restricted once published, defenders face a narrowing window to prevent breaches. However, organizations can also leverage these identical systems to detect and patch weaknesses. — The Decoder</description><category>Research</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>Nous Research fine-tunes Alibaba&apos;s Qwen model to strip political bias</title><link>https://the-decoder.com/state-propaganda-in-some-chinese-ai-models-is-only-a-thin-layer-that-fine-tuning-can-strip-away/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n91</guid><pubDate>Sat, 07 Nov 2026 12:32:45 GMT</pubDate><description>Nous Research retrained Alibaba&apos;s Qwen3.8-27B on 53 million tokens to produce Arete-27B, stripping out Chinese state propaganda. The model&apos;s propaganda rate dropped from 22 percent to under one percent while maintaining its general capabilities. The team found that the political bias is only a thin layer that can be removed through fine-tuning. — The Decoder</description><category>Models &amp; releases</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>AI agents fail benchmark for open scientific research questions</title><link>https://the-decoder.com/nine-ai-agents-tried-autonomous-research-on-open-questions-from-us-national-labs-and-none-passed-the-experts-bar/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n92</guid><pubDate>Sat, 07 Nov 2026 11:13:42 GMT</pubDate><description>GPT-6 Astra achieved the best results among the tested models, but still fell short of the required threshold. — The Decoder</description><category>Research</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>Google develops AI to detect high blood pressure using standard smartwatches</title><link>https://the-decoder.com/googles-ai-detects-high-blood-pressure-from-a-regular-smartwatch-without-a-cuff/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n93</guid><pubDate>Sat, 07 Nov 2026 09:54:09 GMT</pubDate><description>Google Research has developed an AI model that identifies signs of high blood pressure using standard smartwatch pulse and motion sensors without requiring a cuff. The feature has been integrated into the Google Health app to encourage users to get checked, though it does not serve as an official diagnosis. — The Decoder</description><category>Research</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>OpenAI Secures $11 Billion in Funding at $1.25 Trillion Valuation</title><link>https://the-decoder.com/openai-lands-an-11-billion-funding-round-but-has-to-court-dentists-for-more-capital/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n94</guid><pubDate>Sat, 07 Nov 2026 09:45:23 GMT</pubDate><description>OpenAI has raised an initial $11 billion in a funding round that values the company at $1.25 trillion and could expand to $20 billion. Backers include firms such as SoftBank, Sequoia Capital, and a consortium from the United Arab Emirates, alongside efforts through investment banks to enlist wealthy individual investors. The company scaled down its initial target of $30 billion at a $1.4 trillion valuation following a market selloff in the AI sector. — The Decoder</description><category>Business</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>Google acquires unfinished Wyoming data center project from Crusoe</title><link>https://techcrunch.com/2026/11/06/crusoe-sells-troubled-wyoming-data-center-project-to-google/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n22</guid><pubDate>Sat, 07 Nov 2026 01:40:28 GMT</pubDate><description>Crusoe has sold an incomplete data center project in Wyoming to Google, which plans to invest $17 billion to develop the site. Google assumed control of the facility after previously raising concerns about project costs and construction timelines. The transaction highlights the growing hurdles independent developers encounter when securing power and equipment for large-scale AI infrastructure. — TechCrunch</description><category>Business</category><source url="https://techcrunch.com/category/artificial-intelligence/">TechCrunch</source></item>
<item><title>China Reportedly Recruited Tech Executive to Spy on Nvidia CEO Jensen Huang</title><link>https://www.theverge.com/tech/1029338/nvidia-ceo-jensen-huang-china-spy</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n42</guid><pubDate>Fri, 06 Nov 2026 23:00:44 GMT</pubDate><description>Chinese intelligence enlisted Megaspeed executive Alice Huang to spy on Nvidia CEO Jensen Huang. The Ministry of State Security sought to obtain information regarding American AI chips through the operation. Officials also aimed to potentially influence Nvidia to make its most advanced processors available to China. — The Verge</description><category>Policy &amp; safety</category><source url="https://www.theverge.com/ai-artificial-intelligence">The Verge</source></item>
<item><title>Ricursive Intelligence releases Ripple-65B Preview open coding model</title><link>https://the-decoder.com/ripple-65b-preview-delivers-strong-agentic-coding-results-as-a-fully-apache-2-0-open-us-model/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n95</guid><pubDate>Fri, 06 Nov 2026 19:47:04 GMT</pubDate><description>Ricursive Intelligence has launched Ripple-65B Preview, a 65-billion-parameter open-weight model trained on 12 trillion tokens at a development cost of $32 million. Built with a dense Transformer architecture optimized for Nvidia hardware, it is available on Hugging Face under the Apache 2.0 license. — The Decoder</description><category>Models &amp; releases</category><source url="https://the-decoder.com">The Decoder</source></item>
<item><title>Meta launches Spaces to organize tasks and chats with Muse agents</title><link>https://techcrunch.com/2026/11/06/meta-introduces-spaces-a-new-way-to-collaborate-with-your-muse-agent/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-08/#n25</guid><pubDate>Fri, 06 Nov 2026 18:44:20 GMT</pubDate><description>Meta has introduced Spaces, a feature enabling users to organize dedicated breakout conversations with its Muse AI agents around specific projects or topics. Users can describe an idea in the chat to create a specialized planner or tracker in the sidebar. The feature also supports integrations with third-party apps and data. — TechCrunch</description><category>Tools &amp; products</category><source url="https://techcrunch.com/category/artificial-intelligence/">TechCrunch</source></item>
<item><title>Palo Alto Networks in Talks to Acquire Clutch Security for $500 Million</title><link>https://techcrunch.com/2026/11/05/palo-alto-networks-in-talks-to-buy-ai-agent-security-startup-clutch-for-500m/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n32</guid><pubDate>Fri, 06 Nov 2026 06:28:32 GMT</pubDate><description>Palo Alto Networks is negotiating to acquire Clutch Security, a startup that secures AI agents&apos; access to enterprise systems, for $500 million in cash plus milestone payouts. Clutch is approaching $10 million in annual recurring revenue and was valued at $85 million in January 2025. If completed, the deal would be Palo Alto Networks&apos; seventh acquisition this year as it responds to rising AI-driven cyber threats. — TechCrunch</description><category>Business</category><source url="https://techcrunch.com/category/artificial-intelligence/">TechCrunch</source></item>
<item><title>Open ASR Leaderboard Adds Egyptian and Nigerian English Accents</title><link>https://huggingface.co/blog/open-asr-leaderboard-liva</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n17</guid><pubDate>Fri, 06 Nov 2026 00:00:00 GMT</pubDate><description>The Open ASR Leaderboard has partnered with Liva AI to add private evaluation datasets featuring Egyptian and Nigerian English accents. The update aims to improve accent representation and increase trust in speech recognition benchmarks by preventing over-optimization on public data. The new data also enables a more detailed performance breakdown based on Kachru&apos;s Three Circles model. — Hugging Face</description><category>Arab world</category><source url="https://huggingface.co/blog">Hugging Face</source></item>
<item><title>NVIDIA Releases Four Open Cosmos 3 Models for Interactive World Simulation</title><link>https://huggingface.co/blog/nvidia/cosmos-interactive-world-simulation</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n18</guid><pubDate>Thu, 05 Nov 2026 19:35:16 GMT</pubDate><description>The models are designed for interactive world simulation, generating multi-sensor driving scenes and visual predictions for autonomous vehicles and robotics. They are available for both commercial and non-commercial use under the OpenMDW1.1 license. — Hugging Face</description><category>Models &amp; releases</category><source url="https://huggingface.co/blog">Hugging Face</source></item>
<item><title>Google DeepMind Introduces Gemini Omni 1.2 Flash for Video Generation</title><link>https://deepmind.google/blog/gemini-omni-12-flash-fast-and-cost-effective-with-longer-clips/</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n15</guid><pubDate>Thu, 05 Nov 2026 17:05:16 GMT</pubDate><description>Google DeepMind has introduced Gemini Omni 1.2 Flash, which brings native 15-second video generation and editing capabilities. — Google DeepMind</description><category>Models &amp; releases</category><source url="https://deepmind.google">Google DeepMind</source></item>
<item><title>OpenAI Introduces GPT-6 Cyber for Vulnerability Detection</title><link>https://openai.com/index/introducing-gpt-6-cyber</link><guid isPermaLink="true">https://scouthub.dev/en/news/2026-11-07/#n4</guid><pubDate>Thu, 05 Nov 2026 09:00:00 GMT</pubDate><description>OpenAI has introduced GPT-6 Cyber, a cybersecurity model built to identify exploitable vulnerabilities and generate fixes. The model is trained to remain strictly within its authorized scope. — OpenAI</description><category>Models &amp; releases</category><source url="https://openai.com">OpenAI</source></item>
</channel></rss>