News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
bluesky bluesky-ai | 1mo | www.selc.org/news/were-su... | www.selc.org/news/were-su... | ↗Not briefed |
bluesky bluesky-ai | 1mo | www.groene.nl/artikel/plop... (Gratis als het goed is) | www.groene.nl/artikel/plop... (Gratis als het goed is) | ↗Not briefed |
arxiv arxiv-ai | 1mo | Planned Test-Time Scaling with Coordinated Reasoning Paths | Test-time scaling with parallel branches is widely adopted to improve performance on challenging reasoning tasks. The predominant appr… | ↗Not briefed |
rss techmeme | 1mo | The US criticizes Australia's digital duty-of-care proposal, which has a provision that would require social media platforms to let users opt out of algorithms (Angus Whitley/Bloomberg) | Angus Whitley / Bloomberg: The US criticizes Australia's digital duty-of-care proposal, which has a provision that would require socia… | ↗Not briefed |
bluesky bluesky-ai | 1mo | SoftBank Group has drawn over $20 billion of preliminary demand for what could be one of the biggest junk bond deals ever, as investors rus… | SoftBank Group has drawn over $20 billion of preliminary demand for what could be one of the biggest junk bond deals ever, as investor… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Neither Silence nor Overlap Is Failure: Intent-Conditioned Evaluation of Turn-Taking in Full-Duplex Spoken Dialogue Models | Benchmarks for full-duplex spoken dialogue models score turn-taking with binary fixed-window rules that reward immediate response or s… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Attention Routing Stabilizes Early: Working-Set Inference for Recurrent Language Models | Recurrent language models repeatedly apply shared network blocks to refine latent representations, but standard inference recomputes g… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Geometry-Conditioned Visual Place Recognition in Natural Environments | Visual Place Recognition (VPR) in natural environments remains challenging due to repetitive vegetation, sparse distinctive landmarks,… | ↗Not briefed |
rss bing-news-ai | 1mo | Trump wants to rebrand artificial intelligence as 'super intelligence' | President Trump said at the United National that he is directing "all United States documents" to refer to A.I. as "super intelligence… | ↗Not briefed |
rss marktechpost | 1mo | OpenAI Releases GPT-6 Sol and Luna: 50% Cheaper API Pricing and Benchmarks🔧 ChatGPT | OpenAI has released GPT-6 Sol and GPT-6 Luna, 2 lower-cost models trained with methods similar to GPT-6 Astra. Sol costs $2/$10 and Lu… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Hier ein paar zuletzt häufig geteilte #News: Amoklauf an Schule in Tumbler Ridge: Auch kanadische Provinz klagt gegen OpenAI Gleich noch … | Hier ein paar zuletzt häufig geteilte #News: Amoklauf an Schule in Tumbler Ridge: Auch kanadische Provinz klagt gegen OpenAI Gleich… | ↗Not briefed |
hn hn-ai | 1mo | Making Linux prove a governed action before execution | — | ↗Not briefed |
rss techmeme | 1mo | Patreon co-founder Sam Yam says he is joining OpenAI to lead Creator Product, alongside Patreon's Head of Product Drew Rowny and Head of Engineering Shannon Ma (Sam Yam/@samyamiam) | Sam Yam / @samyamiam: Patreon co-founder Sam Yam says he is joining OpenAI to lead Creator Product, alongside Patreon's Head of Produc… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Het sensationele beursdebuut van Anthropic (Claude) vertoont alle kenmerken van een financiële bubbel. De vraag is niet zozeer of, maar wan… | Het sensationele beursdebuut van Anthropic (Claude) vertoont alle kenmerken van een financiële bubbel. De vraag is niet zozeer of, maa… | ↗Not briefed |
hn hn-ai | 1mo | Fpp – a utility to pick file paths out of command output | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | It's perfect in a way that a compelling piece of LLM art has an uncertain and distributed provenance, only knowable by scraping the depths … | It's perfect in a way that a compelling piece of LLM art has an uncertain and distributed provenance, only knowable by scraping the de… | ↗Not briefed |
hn hn-ai | 1mo | Pg_chdb, fast imports from object storage to Postgres using COPY | — | ↗Not briefed |
rss techmeme | 1mo | Sources and Nscale's US SEC filings: ByteDance accounted for nearly 75% of Nscale's sales in 2025 and used Nscale's facility in Norway to access Nvidia AI chips (Financial Times) | Financial Times: Sources and Nscale's US SEC filings: ByteDance accounted for nearly 75% of Nscale's sales in 2025 and used Nscale's f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Anomaly-Free Self-Optimization via AUC Bounds | Anomalies are rare, and anomalous data are often unavailable during development, making it difficult to determine which anomaly detect… | ↗Not briefed |
rss techcrunch-ai | 1mo | “We’re already fighting yesterday’s battle”: Greece’s prime minister gets candid about AI | Most leaders on a trade mission stick to the pitch, but when I interviewed Greek Prime Minister Kyriakos Mitsotakis this week, he also… | ↗Not briefed |
bluesky bluesky-ai | 1mo | when i say hey chat PLEASE read that as a streamer talking to their audience and NOT me asking a fucking llm | when i say hey chat PLEASE read that as a streamer talking to their audience and NOT me asking a fucking llm | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ok. Question for fellow AI-knowers: do we even think the frontier model series names actually reflect differences in terms of what the mode… | Ok. Question for fellow AI-knowers: do we even think the frontier model series names actually reflect differences in terms of what the… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Automated Extraction of Records of Processing Activities (RoPA) Using Hybrid RAG and Locally Deployed Large Language Models | Vietnam's Personal Data Protection Law (Law No. 91/2025/QH15) and Decree No. 356/2025/ND-CP, effective January 1, 2026, require organi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Quantization-Robust Unlearning through the Lens of Retain-Forget Loss Landscapes Interaction | Unlearning ensures LLM compliance by removing the influence of private or copyrighted training data. However, since LLM models typical… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Constraint-Driven Context Engineering: Designing Domain Interfaces for AI Systems | Generative AI systems are increasingly deployed to address domain problems. These systems operate under technical, regulatory, institu… | ↗Not briefed |
hn hn-ai | 1mo | Claude Pop – I'm Upping My P(Doom) [video] | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Guides That Cause Actions: An Offline Study of Guide-Action Mutual Reinforcement in Multimodal Web Agents🔧 GPT-5 | Web agents are usually evaluated in live environments, where environment state and judge models drift between runs, so the same checkp… | ↗Not briefed |
bluesky bluesky-ai | 1mo | showed this to my mom and while the bits went over her head she was appropriately horrified that an llm can make this from scratch | showed this to my mom and while the bits went over her head she was appropriately horrified that an llm can make this from scratch | ↗Not briefed |
hn hn-ai | 1mo | TurboKain – A 1.2MB native radio astronomy pipeline that slices 17GB in 2.3s | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Honestly my preferred solution (mind you, my experience is junior faculty in the philosophy dept for three years and absolutely no administ… | Honestly my preferred solution (mind you, my experience is junior faculty in the philosophy dept for three years and absolutely no adm… | ↗Not briefed |
hn hn-ai | 1mo | Choice of over 600 LLMs across 9 providers | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | MolDesignBench: Evaluating LLM-based Agent for Scenario-grounded Molecular Design | Real-world molecular design remains challenging for large language model (LLM)-based agents. It requires them to interpret design cont… | ↗Not briefed |
bluesky bluesky-ai | 1mo | enjoying this page on waymo's site which, sans context, tells you what to do if the AI model controlling your vehicle gets drunk or uses dr… | enjoying this page on waymo's site which, sans context, tells you what to do if the AI model controlling your vehicle gets drunk or us… | ↗Not briefed |
bluesky bluesky-ai | 1mo | マクロンの提案は、単一の第三極となるAI圏の形成というよりは、米中への依存を軽減し、米中主導のAI管理体制に収斂させず、各国に主権的な意思決定の余地を生み出すために中堅国が開発面で協力する構想という印象。また、「少数の企業がすべての決定権を握る」状況も問題視している。これはAn… | マクロンの提案は、単一の第三極となるAI圏の形成というよりは、米中への依存を軽減し、米中主導のAI管理体制に収斂させず、各国に主権的な意思決定の余地を生み出すために中堅国が開発面で協力する構想という印象。また、「少数の企業がすべての決定権を握る」状況も問題視している。… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Castrag – transcribe and semantically search a large podcast archive | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | I have an art degree and I've never been able to articulate this so succintly and I think the beautiful art in the 4th panel is something a… | I have an art degree and I've never been able to articulate this so succintly and I think the beautiful art in the 4th panel is someth… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Napkin – a scratchpad that holds texts and images for copy | Napkin is a simple scratchpad to hold text and images to copy later or until you don't need them anymore, or when you don't… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Though I suspect the EAs and Anthropic could easily be hacked by power and status, so | Though I suspect the EAs and Anthropic could easily be hacked by power and status, so | ↗Not briefed |
bluesky bluesky-ai | 1mo | watching two alleged professionals have an increasingly bitchy & very passive aggressive fight via email using what’s clearly llm outputs &… | watching two alleged professionals have an increasingly bitchy & very passive aggressive fight via email using what’s clearly llm outp… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Evolving Inspectable O-RAN Slicing xApps with LLMs | Open RAN (O-RAN) slicing xApps must adapt resource allocations to changing channel conditions and traffic demands while meeting servic… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Claude Opus 5.5 veröffentlicht + Österreich setzt NIS2 um + EU diskutiert Haftung für KI-Schäden + High-End-Chips von Qualcomm + Sicherheit… | Claude Opus 5.5 veröffentlicht + Österreich setzt NIS2 um + EU diskutiert Haftung für KI-Schäden + High-End-Chips von Qualcomm + Siche… | ↗Not briefed |
arxiv arxiv-ai | 1mo | CART: Closed-Loop Adaptive Red Teaming for Large Language Models | Automated red teaming often replays a fixed set of prompts, which measures known risks but cannot learn from failures found during tes… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents | Agentic memory systems reuse past experience to improve future performance, yet most existing designs curate memory at write time: onc… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Alignment Inertia: Auditing the Durability of Training Data Influence Through Policy Override Resistance | Platform operators increasingly rely on system prompts and fine-tuning to govern model behavior, yet it remains unclear how reliably t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | (btw in modern chess engines since more'n a decade ago, this is also done partially by classifier model. stockfish can look at a position a… | (btw in modern chess engines since more'n a decade ago, this is also done partially by classifier model. stockfish can look at a posit… | ↗Not briefed |
bluesky bluesky-ai | 1mo | truthout.org/articles/rep... | truthout.org/articles/rep... | ↗Not briefed |
arxiv arxiv-ai | 1mo | Stable Geometry with Divergent Task Evidence for Efficient Long-Horizon Agent Compression | Long horizon agents accumulate growing interaction histories that increase context and inference costs. We find that geometric redunda… | ↗Not briefed |
rss bing-news-ai | 1mo | Trump Wants to Rename AI ‘Super Intelligence’ Because ‘Artificial’ Sounds Fake | President Trump says the U.S. will call AI “super intelligence,” but the term already has a different technical meaning in AI research. | ↗Not briefed |
bluesky bluesky-ai | 1mo | finally, real benchmarks www.assbench.com | finally, real benchmarks www.assbench.com | ↗Not briefed |
bluesky bluesky-ai | 1mo | Posting this feathered Tyrannosaur design of mine on Neatoshop for NO REASON WHATSOEVER #sciart #paleoart #science #dinosaur #BSNM Availab… | Posting this feathered Tyrannosaur design of mine on Neatoshop for NO REASON WHATSOEVER #sciart #paleoart #science #dinosaur #BSNM Av… | ↗Not briefed |