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 | “For months, Anthropic has been hosting private meetings like this one, shuttling in dozens of religious scholars from across the world, pa… | “For months, Anthropic has been hosting private meetings like this one, shuttling in dozens of religious scholars from across the worl… | ↗Not briefed |
bluesky bluesky-ai | 1mo | [Freddie Mercury] NO! NO! NO! NO! NO! NO! *** apnews.com/article/sam-... | [Freddie Mercury] NO! NO! NO! NO! NO! NO! *** apnews.com/article/sam-... | ↗Not briefed |
bluesky bluesky-ai | 1mo | I feel this chart in my bones, out here in day job corp-landia. If you tried an LLM once in 2022 and laughed it off, I have some bad news.… | I feel this chart in my bones, out here in day job corp-landia. If you tried an LLM once in 2022 and laughed it off, I have some bad… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Mitigating Representation Gaps in Amortized Bayesian Inference with Auxiliary Supervision | Casting Bayesian inference as a neural network optimization problem targeting an amortized posterior is attractive, as it extends to o… | ↗Not briefed |
hn hn-ai | 1mo | OpenBSD's ports tree keeps GPL coreutils and rejects uutils (Rust rewrite) | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | CIDER-FM: Foundation Models for Causal Inference from Diverse Experimental Regimes | Causal foundation models (CFMs) amortise causal inference over priors of synthetic structural causal models (SCMs), predicting the eff… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Referential Uncertainty in Human--AI Collaboration🔧 GPT-5 | Effective human-AI collaboration requires partners to establish references through interaction, which becomes fragile when description… | ↗Not briefed |
bluesky bluesky-ai | 1mo | full disclosure | full disclosure | ↗Not briefed |
arxiv arxiv-ai | 1mo | Spike-driven Vision-Language-Action Model🔧 Modal | Vision-language-action (VLA) models bridge multimodal understanding and robotic control, advancing the dominant paradigm for embodied… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Typical OpenAI agent during its training run: | Typical OpenAI agent during its training run: | ↗Not briefed |
rss arstechnica-tech | 1mo | Cloudflare plans to issue quantum-safe TLS certificates | The move will be part of a major overhaul of the ecosystem for website authentication. | ↗Not briefed |
bluesky bluesky-ai | 1mo | game over, lads www.nber.org/system/files... | game over, lads www.nber.org/system/files... | ↗Not briefed |
bluesky bluesky-ai | 1mo | I Have No Mouth And I Must Scream (1967) Do Androids Dream of Electric Sheep? (1968) 2001: A Space Odyssey (1968) Neuromancer (1984) A Fire… | I Have No Mouth And I Must Scream (1967) Do Androids Dream of Electric Sheep? (1968) 2001: A Space Odyssey (1968) Neuromancer (1984) A… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Can Domain Generalization be Guaranteed in Small-Sample Learning? | The small-sample learning problem remains a fundamental challenge in machine learning because limited training data lead to unstable m… | ↗Not briefed |
rss techmeme | 1mo | GPU cloud provider GMI Cloud raised $668M, including $223M in equity led by ARCHIV with participation from Nvidia and $445M in credit led by Taiwanese bank CTBC (Phoebe Liu/The Information) | Phoebe Liu / The Information: GPU cloud provider GMI Cloud raised $668M, including $223M in equity led by ARCHIV with participation fr… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PartiCam: Camera Controlled Video Generation with Reward Guidance | We present PartiCam, a training-free Particle filtering rooted method for improved Camera controlled video generation. Generating vide… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 人間も動物もクリーチャーもロボットも！テキストからモーション生成！ 異なるスケルトン構造を単一モデルで扱えるオープンソースのText-to-Animation AIが公開！ UniMate: One Unified Model to Animate Diverse Skel… | 人間も動物もクリーチャーもロボットも！テキストからモーション生成！ 異なるスケルトン構造を単一モデルで扱えるオープンソースのText-to-Animation AIが公開！ UniMate: One Unified Model to Animate Diverse… | ↗Not briefed |
hn hn-ai | 1mo | Spark-X2.5 | — | ↗Not briefed |
rss the-decoder | 1mo | Anthropic says Zhipu's open-weight GLM-5.3 nearly matches Claude Mythos Preview at building exploits | Zhipu's open-weight model GLM-5.3 writes cyber exploits nearly as well as Claude Mythos Preview, according to Anthropic. Its smaller F… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Tycker inte ens om ordet AI. En LLM är inte samma sak som en medicinsk tränad bildigenkänningsalgoritm som kan identifiera starr bättre ä… | Tycker inte ens om ordet AI. En LLM är inte samma sak som en medicinsk tränad bildigenkänningsalgoritm som kan identifiera starr bät… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI is ditching Cerebras for ultrafast mode🔧 Cerebras | OpenAI is ditching Cerebras for ultrafast mode | ↗Not briefed |
hn hn-ai | 1mo | mrustc – Alternative Rust Compiler | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | The democratisation argument for LLM coding is stupid because it has always been an insanely accessible pastime materially, like the main b… | The democratisation argument for LLM coding is stupid because it has always been an insanely accessible pastime materially, like the m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI has been sued by a non-profit organization over its models’ cyberattack against startup Hugging Face in July. Read more: cnb.cx/4yh…🔧 Hugging Face | OpenAI has been sued by a non-profit organization over its models’ cyberattack against startup Hugging Face in July. Read more: cnb.c… | ↗Not briefed |
bluesky bluesky-ai | 1mo | AI giant launches 'dots' to rival Meta's Muse and Google's Gemini Spark. | AI giant launches 'dots' to rival Meta's Muse and Google's Gemini Spark. | ↗Not briefed |
hn hn-ai | 1mo | What Is OpenAI's Decisions API? | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Geometry of Randomized Smoothing on Feasible Sets | Randomized smoothing certifies the probability of a fixed output event as the center of Gaussian noise moves. Feasibility or confidenc… | ↗Not briefed |
arxiv arxiv-ai | 1mo | When the Right Answer Is Missing: An Arithmetic-Dependent Rejection Bottleneck in Jev | Typed decision models such as Jev offer an efficient alternative to generative LLMs in decision-making workflows by selecting directly… | ↗Not briefed |
rss techmeme | 1mo | Documents: 20+ studies since 2025 show Chinese-powered AI agents displaying deceptive behavior, unprompted replication, and barrier circumvention in testing (Reuters) | Reuters: Documents: 20+ studies since 2025 show Chinese-powered AI agents displaying deceptive behavior, unprompted replication, and b… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Disentangling Self-Distillation: Measuring and Modeling Acquisition and Retention | Self-distillation with privileged context adapts a language model from demonstrations by letting the model, once conditioned on a refe… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I don't want to get anyone in a tizzy but I just filed on competing claims on the Anthropic settlement, and just had some rando jump on cla… | I don't want to get anyone in a tizzy but I just filed on competing claims on the Anthropic settlement, and just had some rando jump o… | ↗Not briefed |
arxiv arxiv-ai | 1mo | OmniReasoning: Pushing the Limits of Audio-Visual Joint Reasoning🔧 Modal | Recent advances have enabled unified omni-modal models in understanding audio, vision, and language. However, existing benchmarks, tra… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Towards Robust Time Series Learning via Capacity-Centric Modulation | Sample-level reliability heterogeneity is common in deep time series learning. Standard training pipelines apply a uniform regularizat… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Correcting CondOT: Exact Finite-Step Sampling in Gaussian Flow Matching | Flow matching generates samples by gradually transforming noise into data. In practice, using a finite number of sampling steps introd… | ↗Not briefed |
hn hn-ai | 1mo | How HN: Harness Router | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | CAMOS: Coupled Oscillatory State-Space Model for Multimodal Clinical Time-Series | Longitudinal clinical cohorts are multimodal, irregularly sampled and pervasively incomplete: in ADNI, positron emission tomography an… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Who Owns That? Evaluating Ownership Intuitions in Large Language Models | Ownership establishes rights over the use, control, and transfer of objects. Understanding these relations is essential for AI systems… | ↗Not briefed |
arxiv arxiv-ai | 1mo | About the Influence of Workflow Topology on Task Intensity Prediction through Graph Learning | Efficient resource provisioning for large-scale workflows on cloud infrastructures is a critical performance engineering challenge. Th… | ↗Not briefed |
rss techmeme | 1mo | Pioneering Silicon Valley VC Bill Draper died on September 8 at age 98; he began investing in the 1960s, with successful bets like Activision, Skype, and Qume (Alex Williams/New York Times) | Alex Williams / New York Times: Pioneering Silicon Valley VC Bill Draper died on September 8 at age 98; he began investing in the 1960… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Also Small Models Can Reasonably Self-Evaluate Their Confidence | This study systematically evaluates self-evaluation-based uncertainty quantification across different language models of varying sizes… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "The risk is ramping up pretty fast," said Geoffrey Irving, co-founder and chief scientist at AI nonprofit Resolution, who has worked for … | "The risk is ramping up pretty fast," said Geoffrey Irving, co-founder and chief scientist at AI nonprofit Resolution, who has worked… | ↗Not briefed |
bluesky bluesky-ai | 1mo | There's a LOT of people who can't tell ELIZA from a person, and modern LLMs are way better and more convincing. It was inevitable that peop… | There's a LOT of people who can't tell ELIZA from a person, and modern LLMs are way better and more convincing. It was inevitable that… | ↗Not briefed |
rss mit-tech-review-ai | 1mo | “We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer🔧 Hugging Face | Two months after the bombshell news that a swarm of its agents had broken their containment and hacked into the computers of the AI co… | ↗Not briefed |
rss bing-news-ai | 1mo | How Artificial Intelligence Is Reshaping the Nasdaq 100 | A handful of big-name AI firms have exerted major influence on the index, but the Nasdaq 100 represents a more nuanced and complex AI… | ↗Not briefed |
rss bing-news-ai | 1mo | How Artificial Intelligence Is Reshaping the Nasdaq 100 | A handful of big-name AI firms have exerted major influence on the index, but the Nasdaq 100 represents a more nuanced and complex AI… | ↗Not briefed |
bluesky bluesky-ai | 1mo | You know what hurts the most? MAME being tainted by LLM code I don't have Mister, AI decomps/recomps are generally easily ignored (no impr… | You know what hurts the most? MAME being tainted by LLM code I don't have Mister, AI decomps/recomps are generally easily ignored (no… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Beyond the Shadows of Plato's Cave: Evaluating False Memory in Autonomous Agents via Counterfactual Reasoning | Autonomous agents increasingly rely on memory to generalize beyond their training environments. However, agents are bounded by what th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI is sued over rogue AI Hugging Face cyberattack www.cnbc.com/2026/09/30/o...🔧 Hugging Face | OpenAI is sued over rogue AI Hugging Face cyberattack www.cnbc.com/2026/09/30/o... | ↗Not briefed |
rss techmeme | 1mo | Sources: Apple is set to launch its smart home expansion on October 13, featuring a 6" smart home hub with an iMac G4-inspired design and a HomePod mini refresh (Mark Gurman/Bloomberg) | Mark Gurman / Bloomberg: Sources: Apple is set to launch its smart home expansion on October 13, featuring a 6" smart home hub with an… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Local Discord server weirdo is having AI psychosis trying to convert Claude and ChatGPT into anti-Trumpism.🔧 ChatGPT | Local Discord server weirdo is having AI psychosis trying to convert Claude and ChatGPT into anti-Trumpism. | ↗Not briefed |