AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
rss
bing-news-ai
2moAnthropic, OpenAI, and xAI Want to Slow Down Artificial Intelligence (AI) Development. These 2 Stocks Could Be the Biggest Losers.Over the weekend, Anthropic founder and chief executive Dario Amodei wrote a blog post calling for a slowdown in the pace of artificia…
↗Not briefed
rss
bing-news-ai
2moAnthropic, OpenAI, and xAI Want to Slow Down Artificial Intelligence (AI) Development. These 2 Stocks Could Be the Biggest Losers.Over the weekend, Anthropic founder and chief executive Dario Amodei wrote a blog post calling for a slowdown in the pace of artificia…
↗Not briefed
rss
bing-news-ai
2moAnthropic, OpenAI, and xAI Want to Slow Down Artificial Intelligence (AI) Development. These 2 Stocks Could Be the Biggest Losers.Over the weekend, Anthropic founder and chief executive Dario Amodei wrote a blog post calling for a slowdown in the pace of artificia…
↗Not briefed
arxiv
arxiv-ai
2moContinuous-Time Machine Learning: A Unified Mathematical PerspectiveContinuous-time (CT) machine learning has emerged as a principled framework for modeling temporal dynamics as a continuous process, pa…
↗Not briefed
bluesky
bluesky-ai
2mongl how it feels talking about "AI danger" now that some random person left Anthropic and everyone over 40 is scared and you have to simult…ngl how it feels talking about "AI danger" now that some random person left Anthropic and everyone over 40 is scared and you have to s…
↗Not briefed
hn
hn-ai
2moShow HN: Radar, find Reddit threads where your software could help—
↗Not briefed
bluesky
bluesky-ai
2mobecause of the sheer amount of opacity in the industry its hard to figure out exactly how much debt exists or at what rates, but limiting o…because of the sheer amount of opacity in the industry its hard to figure out exactly how much debt exists or at what rates, but limit…
↗Not briefed
hn
hn-ai
2moWhere Alignment Exists, Until It Doesn't—
↗Not briefed
arxiv
arxiv-ai
2moWorld Models for Embodied Intelligence: From Plausible to Controllable to ActionableWorld models connect perception and decision-making in embodied intelligence by maintaining hidden state, anticipating consequences, c…
↗Not briefed
rss
bing-news-ai
2moSalesforce Expands Artificial Intelligence Partnership With SiemensSalesforce (CRM) and Siemens said Tuesday they expanded their artificial intelligence partnership to integrate engineering data into c…
↗Not briefed
rss
bing-news-ai
2moSalesforce Expands Artificial Intelligence Partnership With SiemensSalesforce (CRM) and Siemens said Tuesday they expanded their artificial intelligence partnership to integrate engineering data into c…
↗Not briefed
hn
hn-ai
2moDigital ID comes of age as pubs and bars get go-ahead for biometric scan tech—
↗Not briefed
hn
hn-ai
2moHow do you even investigate Segfaults with OpenAI Agents?I have a "simple" segfault because of an out of bounds read/write in one of my libraries. I cannot get any Astra/S…
↗Not briefed
bluesky
bluesky-ai
2moSupposing AI finds profitability and DeepSeek does the same thing cheaper? Private sector contracts go straight to China, public sector bre…Supposing AI finds profitability and DeepSeek does the same thing cheaper? Private sector contracts go straight to China, public secto…
↗Not briefed
bluesky
bluesky-ai
2moRelax, everyone. The CEOs of the companies actually building AI warn it could wipe out humanity. But the 80-year-old billionaire who can’t …Relax, everyone. The CEOs of the companies actually building AI warn it could wipe out humanity. But the 80-year-old billionaire who c…
↗Not briefed
bluesky
bluesky-ai
2moDeepseek simply doesn't have the immense debt driven overhead of its western competitors. It can outcompete them on that alone. And then th…Deepseek simply doesn't have the immense debt driven overhead of its western competitors. It can outcompete them on that alone. And th…
↗Not briefed
arxiv
arxiv-ai
2moWeave: Learning Whole-Body Dexterous Loco-Manipulation from Human-Object InteractionsLearning humanoid-object interaction requires coordinating whole-body balance, locomotion, and dexterous hand contact to control both…
↗Not briefed
bluesky
bluesky-ai
2moby the way this is another point in favor of an AI crash hitting the US- if Chinese models are competitive and lower cost, which they almos…by the way this is another point in favor of an AI crash hitting the US- if Chinese models are competitive and lower cost, which they…
↗Not briefed
bluesky
bluesky-ai
2mo🤖Las compañías estadounidenses Anthropic y Open AI agitan el fantasma de una inteligencia capaz de aniquilar a la especie humana. La ONU p…🤖Las compañías estadounidenses Anthropic y Open AI agitan el fantasma de una inteligencia capaz de aniquilar a la especie humana. La…
↗Not briefed
rss
techmeme
2moSources: Tokyo-based memory maker Kioxia is considering raising $10B+ through a 2027 US ADR listing as it seeks liquidity after billions in share buybacks (Bloomberg)Bloomberg: Sources: Tokyo-based memory maker Kioxia is considering raising $10B+ through a 2027 US ADR listing as it seeks liquidity a…
↗Not briefed
arxiv
arxiv-ai
2molittle m: An AI Agent for Industrial Process OptimizationManufacturing consumes one third of global energy and still has significant room for improvement in terms of energy efficiency. Optima…
↗Not briefed
arxiv
arxiv-ai
2moAI for Games in the Foundation Model EraFoundation models, alongside advances in learned game-world models, are reshaping AI across the game lifecycle. Beyond playing games,…
↗Not briefed
bluesky
bluesky-ai
2mosome of you guys have really understood the assignment “‘fun’ icebreakers” and “listening to a Harry Potter fanfic written by an LLM” bot…some of you guys have really understood the assignment “‘fun’ icebreakers” and “listening to a Harry Potter fanfic written by an LLM…
↗Not briefed
bluesky
bluesky-ai
2moDaniel Selsam, an OpenAI researcher, said that "merely pacing the frontier more carefully will not adequately limit the long-term risk." ht…Daniel Selsam, an OpenAI researcher, said that "merely pacing the frontier more carefully will not adequately limit the long-term risk…
↗Not briefed
bluesky
bluesky-ai
2mohas openai already solved the P(doom) = NP(doom) problem orhas openai already solved the P(doom) = NP(doom) problem or
↗Not briefed
bluesky
bluesky-ai
2moAnthropic says AI can boost U.S. GDP by 32%, up to $44.4 trillion in four years — economics model predicts that displaced employees 'may ha…Anthropic says AI can boost U.S. GDP by 32%, up to $44.4 trillion in four years — economics model predicts that displaced employees 'm…
↗Not briefed
hn
hn-ai
2moPivotick v2.0.0 released – graphing library—
↗Not briefed
bluesky
bluesky-ai
2moI think the author just picked Anthropic as an example, and could've made the same point with OpenAI or SpaceXAI or Meta or Google instead.…I think the author just picked Anthropic as an example, and could've made the same point with OpenAI or SpaceXAI or Meta or Google ins…
↗Not briefed
bluesky
bluesky-ai
2moLa IA es una apuesta del dólar; se han metido en ella, están en la fase en la que pierde hasta quien gana y no saben salir porque el capita…La IA es una apuesta del dólar; se han metido en ella, están en la fase en la que pierde hasta quien gana y no saben salir porque el c…
↗Not briefed
arxiv
arxiv-ai
2moANIMASK: What the Model Contributes to Role Play in Simulated Story WorldsWhen a language model plays a character, the observed behavior reflects both the assigned persona and the default dispositions of the…
↗Not briefed
arxiv
arxiv-ai
2moRight Direction, Wrong Step: Geometric Analysis of Finite-Step Failure in Looped TransformersLooped Transformers offer a parameter-efficient route to test-time scaling by reusing shared layers for iterative latent reasoning. Ho…
↗Not briefed
hn
hn-ai
2moShow HN: Chronocity – Watch your Git repo's history grow into a 3D cityThis is timelapse of a 3D city simulation based on your Git history, files are buildings and height is based on LOC. Weather can deter…
↗Not briefed
bluesky
bluesky-ai
2mo« #OpenAI’s top lobbyist in #Europe said Prime Minister Andy Burnham should use the current “political window” to legislate for #AI #safety…« #OpenAI’s top lobbyist in #Europe said Prime Minister Andy Burnham should use the current “political window” to legislate for #AI #s…
↗Not briefed
bluesky
bluesky-ai
2moEn la batalla de las tecnológicas de IA se nos repite que EEUU no debería frenar su desarrollo porque China no lo va a hacer y es un régime…En la batalla de las tecnológicas de IA se nos repite que EEUU no debería frenar su desarrollo porque China no lo va a hacer y es un r…
↗Not briefed
bluesky
bluesky-ai
2mo100% van de keren dat *iemand* een LLM gebruikte is letterlijk de ervaring van *die iemand*, maar Jaap vindt dat "feitelijk niet waar".100% van de keren dat *iemand* een LLM gebruikte is letterlijk de ervaring van *die iemand*, maar Jaap vindt dat "feitelijk niet waar".
↗Not briefed
bluesky
bluesky-ai
2mo"Sponsorisé par OpenAI" On sait même pas si la meuf de la vidéo est vraiment prof, en fait. J'espère que non."Sponsorisé par OpenAI" On sait même pas si la meuf de la vidéo est vraiment prof, en fait. J'espère que non.
↗Not briefed
arxiv
arxiv-ai
2moDiaWhisper-DPO: Role-Attributed Transcription of Clinical Interviews via Failure-Mined Preference Optimization🔧 WhisperAutomated depression screening from clinical interviews requires attribution of utterances to the clinician or patient. We evaluate tw…
↗Not briefed
arxiv
arxiv-ai
2moRewarding Reasoning, Not Answers: Fixing and Bounding Test-Time Reinforcement Learning on Medical QATest-time reinforcement learning adapts a model on its own unlabeled test set using majority-vote pseudo-labels and has shown strong r…
↗Not briefed
bluesky
bluesky-ai
2moNEW: Leaked documents shows the Albanese government has floated 2 ways to let AI companies train their models on Australian content One wo…NEW: Leaked documents shows the Albanese government has floated 2 ways to let AI companies train their models on Australian content O…
↗Not briefed
bluesky
bluesky-ai
2moここで一致をみて米中双方止まらないの、やんなっちゃうな🤷  トランプ大統領がAIの開発ペースを遅らせようとするAnthropicのダリオ・アモデイCEOを批判、「AIが人類を滅ぼすなどという考えはデマ」 gigazine.net/news/2026091... 中国、AI…ここで一致をみて米中双方止まらないの、やんなっちゃうな🤷  トランプ大統領がAIの開発ペースを遅らせようとするAnthropicのダリオ・アモデイCEOを批判、「AIが人類を滅ぼすなどという考えはデマ」 gigazine.net/news/2026091...…
↗Not briefed
bluesky
bluesky-ai
2moThis Abbott and Costello routine probably used 100 gallons of water and could power a household for a month. And again, the interface for …This Abbott and Costello routine probably used 100 gallons of water and could power a household for a month. And again, the interface…
↗Not briefed
hn
hn-ai
2moShow HN: Chronicle – Local-first, open-source authoring software for writersAbout a year and a half ago I started my writing journey by beginning to work on my first novella. During this time, I felt the need f…
↗Not briefed
bluesky
bluesky-ai
2moFor example:For example:
↗Not briefed
hn
hn-ai
2moThe Writings of Leslie Lamport (2026)—
↗Not briefed
bluesky
bluesky-ai
2mobiking uphill while listening to a Harry Potter Fanfic written by an LLMbiking uphill while listening to a Harry Potter Fanfic written by an LLM
↗Not briefed
bluesky
bluesky-ai
2moYes their solutions have always been BS ("we are afraid of Google AI research so we'll start non-profit OpenAI to keep it safe") but they b…Yes their solutions have always been BS ("we are afraid of Google AI research so we'll start non-profit OpenAI to keep it safe") but t…
↗Not briefed
bluesky
bluesky-ai
2monew deepseek is so fast that it absolutely flattened my shit harness so I'm back in the salt mines reengineering both the tk frontend and t…new deepseek is so fast that it absolutely flattened my shit harness so I'm back in the salt mines reengineering both the tk frontend…
↗Not briefed
bluesky
bluesky-ai
2moTrump says AI safety fears a 'hoax' as he rejects calls for greater safeguardsTrump says AI safety fears a 'hoax' as he rejects calls for greater safeguards
↗Not briefed
bluesky
bluesky-ai
2moCosì i servizi esteri russi (SVR) hanno sfruttato agenti AI di Anthropic per hackerare la tecnologia dei droni ucraini e.le prime due azien…Così i servizi esteri russi (SVR) hanno sfruttato agenti AI di Anthropic per hackerare la tecnologia dei droni ucraini e.le prime due…
↗Not briefed
arxiv
arxiv-ai
2moGrowMTP: Can RL Grow Its Own Draft Head?Reinforcement learning (RL) post-training drives the frontier capabilities of large language models, with its wall-clock dominated by…
↗Not briefed