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 |
|---|---|---|---|---|
hn hn-ai | 5d | OpenQuantum's DIY Open-Source Magneto-Optical Atomic Trap | — | ↗Not briefed |
bluesky bluesky-ai | 5d | Their Mad Genius Special Boy is needed to keep recirculating the same $20 they keep passing around to keep the AI (grift) business model af… | Their Mad Genius Special Boy is needed to keep recirculating the same $20 they keep passing around to keep the AI (grift) business mod… | ↗Not briefed |
bluesky bluesky-ai | 5d | Met onverzadigbare honger vreet de AI-industrie alles wat los en vast zit, van bossen, chips en boeken tot elektriciteit, water en privégeg… | Met onverzadigbare honger vreet de AI-industrie alles wat los en vast zit, van bossen, chips en boeken tot elektriciteit, water en pri… | ↗Not briefed |
bluesky bluesky-ai | 5d | FINALLY they've figured out how to price cache hits sanely, now if they could just cure themselves of the anthropic disease and stop chargi… | FINALLY they've figured out how to price cache hits sanely, now if they could just cure themselves of the anthropic disease and stop c… | ↗Not briefed |
hn hn-ai | 5d | US and Russia to oppose talks on limiting autonomous weapons | — | ↗Not briefed |
bluesky bluesky-ai | 5d | Okay, folks, that's it. I, a human being, just read (ingested?) 300+ pages of the Anthropic S-1. Tomorrow I'll write about it. Thanks for s… | Okay, folks, that's it. I, a human being, just read (ingested?) 300+ pages of the Anthropic S-1. Tomorrow I'll write about it. Thanks… | ↗Not briefed |
rss techmeme | 5d | Apple launches its Health AI Usage Study, with an OpenEvidence-powered app that answers health questions and can use HealthKit data, for up to 200,000 US users (Ivan Mehta/TechCrunch) | Ivan Mehta / TechCrunch: Apple launches its Health AI Usage Study, with an OpenEvidence-powered app that answers health questions and… | ↗Not briefed |
arxiv arxiv-ai | 5d | A Temporal Convolutional Network with Multi-Head Attention Model: A Surrogate for Parallel DEVS Semiconductor Fab Simulation | Semiconductor factories process many kinds of wafers organized into lots of various sizes and configurations. Different kinds of machi… | ↗Not briefed |
bluesky bluesky-ai | 5d | 700 million sales and hr ghouls pretending they have ai startups just having claude spit out the dumbest shit in defense of this idiot | 700 million sales and hr ghouls pretending they have ai startups just having claude spit out the dumbest shit in defense of this idiot | ↗Not briefed |
arxiv arxiv-ai | 5d | EVIDENT: Evaluating Evidence Grounding in GUI Agents | GUI agents are increasingly evaluated by whether they complete web tasks, but surface success can hide a critical failure: the final a… | ↗Not briefed |
bluesky bluesky-ai | 5d | Legislation that will enable commercially-operated US-owned Pine Gaps in your suburb “If the minister says you're a defence official, well… | Legislation that will enable commercially-operated US-owned Pine Gaps in your suburb “If the minister says you're a defence official,… | ↗Not briefed |
arxiv arxiv-ai | 5d | Agent-Guided Configuration Tuning for SMT Solvers | SMT solvers expose many configuration choices that determine how well their existing algorithms perform on a workload. We present, to… | ↗Not briefed |
arxiv arxiv-ai | 5d | Parallelizable Interior-Point ADMM for Block-Structured Nonconvex Programs🔧 Consensus | Physics simulation is an increasingly important computational primitive for AI, and local--global/ADMM splitting exposes its parallel… | ↗Not briefed |
hn hn-ai | 5d | What music does Gemini Decision 1 like? | — | ↗Not briefed |
arxiv arxiv-ai | 5d | Separating learning from search and feedback in budgeted inverse design of periodic unit cells | To design a periodic material, engineers choose the shape and dimensions of a repeating cell to meet target properties. With few simul… | ↗Not briefed |
bluesky bluesky-ai | 5d | wirae.me/%EC%9D%BC%EA... 부정행위보다 위험한 것 LLM이 야기하는 부정행위에 대한 이야기가 부쩍 많아졌는데, 행정 비용이 들지만 기술적으로 막을 수 있는 부정행위보다 더 위험한 것이 무엇인지에 대해 썼습니다. | wirae.me/%EC%9D%BC%EA... 부정행위보다 위험한 것 LLM이 야기하는 부정행위에 대한 이야기가 부쩍 많아졌는데, 행정 비용이 들지만 기술적으로 막을 수 있는 부정행위보다 더 위험한 것이 무엇인지에 대해 썼습니다. | ↗Not briefed |
rss techmeme | 5d | Alibaba says Qwen4-Max, which isn't open weight, beats Fable 5.1 and Kimi K3.1 Pro in coding benchmarks, and plans to release weights for smaller Qwen4 models (Luz Ding/Bloomberg) | Luz Ding / Bloomberg: Alibaba says Qwen4-Max, which isn't open weight, beats Fable 5.1 and Kimi K3.1 Pro in coding benchmarks, and pla… | ↗Not briefed |
arxiv arxiv-ai | 5d | Beyond Seeing: Fine-Grained Preference Validation and Severity-Aware Alignment for Mitigating Visual Hallucinations in Vision-Language Models | Large vision-language models can produce fluent responses containing objects, attributes, or relations unsupported by the image. We st… | ↗Not briefed |
rss techmeme | 5d | Alibaba releases Qwen4-Max, its new flagship 2.4T-parameter MoE model with a 1M-token context window, saying it beats Qwen3.8-Max with up to ~50% fewer tokens (Qwen) | Qwen: Alibaba releases Qwen4-Max, its new flagship 2.4T-parameter MoE model with a 1M-token context window, saying it beats Qwen3.8-Ma… | ↗Not briefed |
arxiv arxiv-ai | 5d | Improving X-Ray Baggage Instance Segmentation with Shared Point-Based Mask Refinement | Overlapping projections and complex object contours make prohibited items difficult to delineate in X-ray baggage images. This study i… | ↗Not briefed |
arxiv arxiv-ai | 5d | Model Immunization Beyond Condition Numbers: The Importance of Plateau Regions | Harmful fine-tuning can rapidly compromise safety alignment in Large Language Models, underscoring the vulnerability of deployed model… | ↗Not briefed |
arxiv arxiv-ai | 5d | Tabular Foundation Models as Emulators for Forecasting Terrestrial Ecosystem Dynamics | Process-based terrestrial biosphere models such as the Ecosystem Demography (ED) model underpin global carbon budgets, but their cost… | ↗Not briefed |
arxiv arxiv-ai | 5d | On the Applicability of $p$-adic Numbers in Computational Modules for Neural Networks | The applicability of $p$-adic arithmetic as an intermediate computational module in an otherwise unchanged machine-learning pipeline i… | ↗Not briefed |
arxiv arxiv-ai | 5d | ReRead: Wrong Predictions, Useful Representations - Correcting Frozen Robot Policies by Re-reading Their Own Representations | When a pretrained robot policy fails on challenging tasks, common remedies include retraining the backbone, adding an external visual… | ↗Not briefed |
arxiv arxiv-ai | 5d | AdaModalX: Synchronization-Free MFM Inference for Distributed Multimodal Mobile Sensing via Progressive Data Orchestration | Multimodal foundation models (MFMs) support generalizable inference in distributed mobile sensing applications by integrating heteroge… | ↗Not briefed |
rss techmeme | 5d | Sources: Jensen Huang had planned to pull back Nvidia's aggressive financing and told neoclouds to get upfront customer payments; Colette Kress may go further (Wall Street Journal) | Wall Street Journal: Sources: Jensen Huang had planned to pull back Nvidia's aggressive financing and told neoclouds to get upfront cu… | ↗Not briefed |
bluesky bluesky-ai | 5d | Edison Carter (Bill Hader) is one of a dwindling number of human reporters for one of the three ideologically-branded major news outlets, a… | Edison Carter (Bill Hader) is one of a dwindling number of human reporters for one of the three ideologically-branded major news outle… | ↗Not briefed |
bluesky bluesky-ai | 5d | Jensen being reinstated at Nvidia is as stupid as Sam Altman being reinstated at OpenAI, & of course people are calling for it for the same… | Jensen being reinstated at Nvidia is as stupid as Sam Altman being reinstated at OpenAI, & of course people are calling for it for the… | ↗Not briefed |
arxiv arxiv-ai | 5d | Designing Deceptive Model Organisms that Successfully Evade Auditing | AI models can exhibit deceptive behaviors such as strategically underperforming on evaluations (sandbagging), faking alignment, or cov… | ↗Not briefed |
bluesky bluesky-ai | 5d | It is kind of fascinating to see people go off on this website about how "AI can't count the number of Rs in strawberry" while I have an LL… | It is kind of fascinating to see people go off on this website about how "AI can't count the number of Rs in strawberry" while I have… | ↗Not briefed |
arxiv arxiv-ai | 5d | VISTA: A Framework for Vision-Informed Semantic Traffic Awareness | The Internet of Things (IoT) is transforming intelligent transportation systems by connecting roadside cameras, edge computing units,… | ↗Not briefed |
bluesky bluesky-ai | 5d | I think a big difference between me and most tensorfuckers is that I don't come at it from the tech industry so I never had to work with an… | I think a big difference between me and most tensorfuckers is that I don't come at it from the tech industry so I never had to work wi… | ↗Not briefed |
arxiv arxiv-ai | 5d | XMimic and XSoma: Learning Humanoid Motion Tracking and Locomotion in Under a Minute | We present XMimic and XSoma, which learn humanoid motion tracking and locomotion on rough terrain in under a minute of training. Both… | ↗Not briefed |
arxiv arxiv-ai | 5d | Generative AI-Driven Parametric Design Optimization Framework for Vehicle Closure Systems | This study presents a generative AI-driven parametric design optimization framework integrating performance-oriented generation with e… | ↗Not briefed |
rss techmeme | 5d | Mount Thor, which is building cloud infrastructure for agentic AI on Apple silicon, raised $45M in a seed round and a Series A co-led by CRV and Conviction (Bradley C/9to5Mac) | Bradley C / 9to5Mac: Mount Thor, which is building cloud infrastructure for agentic AI on Apple silicon, raised $45M in a seed round a… | ↗Not briefed |
arxiv arxiv-ai | 5d | The Logic Chain Breaking Point: A Controlled-Variable Benchmark for Reasoning Degradation in Open-Weight Language Models | We introduce The Logic Chain Breaking Point, a controlled-variable benchmark for deductive reasoning in open-weight language models. I… | ↗Not briefed |
arxiv arxiv-ai | 5d | Explainable Federated Learning via Global-Local Attribution Alignment | Federated learning enables on-device training without centralizing data, yet existing systems still struggle to provide explanations t… | ↗Not briefed |
bluesky bluesky-ai | 5d | NISA自体は別にいいんだけど、無理してまで投資する若者や、「乗り遅れるな」的に焦ってる人を見ると、NISAの制度に前提として組み込まれている長期投資の方針とか、オルカンとかも(普通に減る時は減る)正真正銘のリスク資産であることとか、ここ数年の上昇幅は概ねAI/LLM関連株が調… | NISA自体は別にいいんだけど、無理してまで投資する若者や、「乗り遅れるな」的に焦ってる人を見ると、NISAの制度に前提として組み込まれている長期投資の方針とか、オルカンとかも(普通に減る時は減る)正真正銘のリスク資産であることとか、ここ数年の上昇幅は概ねAI/LLM… | ↗Not briefed |
arxiv arxiv-ai | 5d | Abstention Direction in Large Language Models: Mechanistic Interpretability of Abstention Across Query Ambiguity, Knowledge Limits, and Safety Alignment | Abstention, the withholding of a response by a large language model (LLM) under uncertainty or safety risks, is essential for trustwor… | ↗Not briefed |
bluesky bluesky-ai | 5d | remember that academic who was surprised to find that people actually liked his emails with occasional typos and mistakes better than the s… | remember that academic who was surprised to find that people actually liked his emails with occasional typos and mistakes better than… | ↗Not briefed |
arxiv arxiv-ai | 5d | Shortcut Learning in Feature-Based Algorithm Selection | Landscape features give learning models information about the objective function for algorithm selection, algorithm configuration, and… | ↗Not briefed |
bluesky bluesky-ai | 5d | (tl:dr they used Claude btw and just trusted that it didn't bullshit anything lmao) why are actual historians/mathematicians/scientists p… | (tl:dr they used Claude btw and just trusted that it didn't bullshit anything lmao) why are actual historians/mathematicians/scienti… | ↗Not briefed |
bluesky bluesky-ai | 5d | And you're encouraging people to trust an LLM agent reading Markdown files off a random github repo with root access to a device in their h… | And you're encouraging people to trust an LLM agent reading Markdown files off a random github repo with root access to a device in th… | ↗Not briefed |
bluesky bluesky-ai | 5d | Asian shares fall before Nvidia earnings, Anthropic IPO test tech rally - www.reuters.com/world/china/... | Asian shares fall before Nvidia earnings, Anthropic IPO test tech rally - www.reuters.com/world/china/... | ↗Not briefed |
bluesky bluesky-ai | 5d | As Democrats debate how to use their new majorities to hold Big Tech accountable, Chuck Schumer's daughter remains a top lobbyist for OpenA… | As Democrats debate how to use their new majorities to hold Big Tech accountable, Chuck Schumer's daughter remains a top lobbyist for… | ↗Not briefed |
bluesky bluesky-ai | 5d | A former xAI engineer was accused of copying Grok trade secrets after selling $7 million in company shares. The lawsuit against OpenAI was … | A former xAI engineer was accused of copying Grok trade secrets after selling $7 million in company shares. The lawsuit against OpenAI… | ↗Not briefed |
arxiv arxiv-ai | 5d | On the Convergence of Continuous-Time AdamW | Continuous-time AdamW provides a framework for understanding whether an optimizer widely used to train large-scale neural networks set… | ↗Not briefed |
hn hn-ai | 5d | Relicense under AGPL-3.0 and add a CLA | — | ↗Not briefed |
bluesky bluesky-ai | 5d | Qwen 4 Max is looking an awful lot like a frontier model * 2.4T MoE * $1.50/$4.50 * attention is getting sophisticated * API serves from S… | Qwen 4 Max is looking an awful lot like a frontier model * 2.4T MoE * $1.50/$4.50 * attention is getting sophisticated * API serves f… | ↗Not briefed |
arxiv arxiv-ai | 5d | iEEG Can Help EEG: Improving Clinical Scalp EEG Analysis through Intracranial Priors🔧 Modal | Scalp electroencephalography (EEG) is a cornerstone of clinical brain monitoring, making the quality of its signal fundamental to down… | ↗Not briefed |