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
bluesky
bluesky-ai
1moSam Altman is a disgusting little twat and should be removed from society www.politico.com/news/2026/10...Sam Altman is a disgusting little twat and should be removed from society www.politico.com/news/2026/10...
↗Not briefed
bluesky
bluesky-ai
1moMaybe an LLM can make that sentence sound like something written by a human who is not in a marketing department.Maybe an LLM can make that sentence sound like something written by a human who is not in a marketing department.
↗Not briefed
hn
hn-ai
1moOpen Source Instinct—
↗Not briefed
bluesky
bluesky-ai
1mo"Asked specifically where he diverges from Anthropic and its CEO Dario Amodei, Altman said 'we believe that the world should accept some ba…"Asked specifically where he diverges from Anthropic and its CEO Dario Amodei, Altman said 'we believe that the world should accept so…
↗Not briefed
arxiv
arxiv-ai
1moAtomic Visual Entailment: Enhancing Zero-Shot Vision-Language Reasoning through Atomic Fact Decomposition and Learned SelectionVisual entailment (VE) asks whether an image supports, contradicts, or leaves undecided a textual hypothesis. Strong results come from…
↗Not briefed
bluesky
bluesky-ai
1moif OpenAI and Anthropic died tomorrow, I don’t think it would have much impact on America’s “competitive edge”. Massive investment would sw…if OpenAI and Anthropic died tomorrow, I don’t think it would have much impact on America’s “competitive edge”. Massive investment wou…
↗Not briefed
bluesky
bluesky-ai
1moOn Sam Altman: “You need to understand that Sam can never be trusted,” Aaron Swartz told one confidante. “He is a sociopath. He would do a…On Sam Altman: “You need to understand that Sam can never be trusted,” Aaron Swartz told one confidante. “He is a sociopath. He would…
↗Not briefed
arxiv
arxiv-ai
1moA Framework for Automated Multi-Source Satellite Data Analytics and LLM-Based Report GenerationThis paper presents the workflow for building an automated ArcGIS Pro tool using ArcPy to extract the Land Surface Temperature (LST) f…
↗Not briefed
bluesky
bluesky-ai
1moI mean, I think this dude just got Claude or Chat to come up with a bad analogy. I'm sure his AI is telling him how clever this retort was …I mean, I think this dude just got Claude or Chat to come up with a bad analogy. I'm sure his AI is telling him how clever this retort…
↗Not briefed
bluesky
bluesky-ai
1mo85.46% on harnessbench for Peritus is apparently meaningless but "heres a shitty claude copy on a cron job" is what gets the hype, the AI c…85.46% on harnessbench for Peritus is apparently meaningless but "heres a shitty claude copy on a cron job" is what gets the hype, the…
↗Not briefed
bluesky
bluesky-ai
1moThe question is not: Can AI have consciousness but how do we establish that. The hard problem. New blog: philosophybytheway.blogspot.com/20…The question is not: Can AI have consciousness but how do we establish that. The hard problem. New blog: philosophybytheway.blogspot.c…
↗Not briefed
arxiv
arxiv-ai
1moUndoBench: Separating Task Competence from Recovery Capability in Tool-Using AI AgentsTool-using AI agents are increasingly deployed across enterprise software systems, yet widely used benchmarks primarily evaluate nomin…
↗Not briefed
bluesky
bluesky-ai
1moでは、紗奈と白夜をどうぞ 五日間の調査で、被験体たちの真実が一つ繋がった。 白夜を含む彼らは、すべて「光の消失」に関わる重要な存在だったのだ。 組織は彼らを隠す必要があった。 紗奈は腹を決めた。 「もっと大きな報道が必要だ」 白夜は頷いた。 明け方の光。 二人は個室を出…では、紗奈と白夜をどうぞ 五日間の調査で、被験体たちの真実が一つ繋がった。 白夜を含む彼らは、すべて「光の消失」に関わる重要な存在だったのだ。 組織は彼らを隠す必要があった。 紗奈は腹を決めた。 「もっと大きな報道が必要だ」 白夜は頷いた。 明け方の光。 二人…
↗Not briefed
rss
simon-willison
1moQwen3.8 27B addition in words—
↗Not briefed
hn
hn-ai
1moEvery Map Everywhere All at Once: exact neutral redistricting maps on a GPU—
↗Not briefed
hn
hn-ai
1moShow HN: EditUI – Cursor/Claude Design-style UI editing directly in the browser—
↗Not briefed
bluesky
bluesky-ai
1mo“Anthropic seeks a Political Programs Lead based in Washington D.C. to build and run Anthropic's presidential engagement program for the 20…“Anthropic seeks a Political Programs Lead based in Washington D.C. to build and run Anthropic's presidential engagement program for t…
↗Not briefed
bluesky
bluesky-ai
1mo“Anthropic is hiring a Political Programs Lead to build and run the programs through which Anthropic, its political action committees, and …“Anthropic is hiring a Political Programs Lead to build and run the programs through which Anthropic, its political action committees,…
↗Not briefed
hn
hn-ai
1moRustQC: Rust reimplementation of dupRadar, RSeQC, Qualimap, preseq, and samtools—
↗Not briefed
hn
hn-ai
1moAn OpenAI agent reached four Australian government systems. Nobody noticed—
↗Not briefed
arxiv
arxiv-ai
1moCausal Lag Structure Discovery in Confounded Time Series via Orthogonalized Adaptive EstimationFinding which variables cause which others in multivariate time series, and at what lags, is central to science and policy, yet existi…
↗Not briefed
bluesky
bluesky-ai
1moThere is an argument that some AI-psychosis-sufferers tend to make which I find very gross, stupid, and bad. It goes something like, “the w…There is an argument that some AI-psychosis-sufferers tend to make which I find very gross, stupid, and bad. It goes something like, “…
↗Not briefed
hn
hn-ai
1moShow HN: Soulor – Rehearse hard conversations with simulated people🔧 ModalI built Soulor independently as a 16y/o high school student. It helps reading social situations with multiple perspectives, and h…
↗Not briefed
bluesky
bluesky-ai
1moThis popped up for me on Reddit. I’d seen it before but I had to watch again, it’s so funny.This popped up for me on Reddit. I’d seen it before but I had to watch again, it’s so funny.
↗Not briefed
bluesky
bluesky-ai
1moI’ve been building a Python-based multi-agent framework with LangGraph, RAG, FastAPI, MCP & tool-based workflows. I’d really appreciate fee…🔧 LangGraphI’ve been building a Python-based multi-agent framework with LangGraph, RAG, FastAPI, MCP & tool-based workflows. I’d really appreciat…
↗Not briefed
arxiv
arxiv-ai
1moDREAM: Dynamic Resolution Assignment For Multimodal Multi-agent DebateMulti-agent debate (MAD) has emerged as an effective paradigm to improve the reasoning capabilities of large language models (LLMs) an…
↗Not briefed
arxiv
arxiv-ai
1moSEA-LM: Egocentric Spatial Audio Understanding for Wearable Microphone ArraysEmbodied, ego-centric intelligence fundamentally requires the ability to comprehend spatial audio within complex environments. While l…
↗Not briefed
arxiv
arxiv-ai
1moKandinsky 6.0 Video: Foundation Models for Synchronized Video and Audio GenerationWe present Kandinsky 6.0 Video, a family of foundation diffusion models for synchronized text-to-audio-video generation, comprising Ka…
↗Not briefed
hn
hn-ai
1moShow HN: C++ LSP for the LeetCode editor and Vim keybindingsPretty self-explanatory tbh. I kept hitting the same friction using LeetCode, the built-in editor has no real language support and non…
↗Not briefed
bluesky
bluesky-ai
1moI don’t think I’ve ever seen someone worse at selling their “revolutionary technology” than this freak.I don’t think I’ve ever seen someone worse at selling their “revolutionary technology” than this freak.
↗Not briefed
bluesky
bluesky-ai
1moNanananananananananananana Deepseek!Nanananananananananananana Deepseek!
↗Not briefed
hn
hn-ai
1moFerguson's Law: Debt Service, Military Spending, and the Limits of Power (2025)—
↗Not briefed
arxiv
arxiv-ai
1moYour Unlearning Gives You Away: Identifying Erased Concepts in Diffusion ModelsExisting attacks on unlearned diffusion models assume that the erased concepts are known in advance and focus on recovering them. In p…
↗Not briefed
hn
hn-ai
1moYuekbox Yue2 Music Generator UI—
↗Not briefed
bluesky
bluesky-ai
1moI am anti gAI use. I have never intentionally logged into an LLM. I am angry daily at my lack of choice & consent to how much gAI exposure…I am anti gAI use. I have never intentionally logged into an LLM. I am angry daily at my lack of choice & consent to how much gAI exp…
↗Not briefed
bluesky
bluesky-ai
1moMcDonald’s doesn’t have to threaten me to eat French fries. Netflix never told me that if I didn’t subscribe, I’d be thrown onto the garba…McDonald’s doesn’t have to threaten me to eat French fries. Netflix never told me that if I didn’t subscribe, I’d be thrown onto the…
↗Not briefed
bluesky
bluesky-ai
1moIf you're doing Engineering right - as distinct from Programming - it's less important whether human or LLM performs the implementation of …If you're doing Engineering right - as distinct from Programming - it's less important whether human or LLM performs the implementatio…
↗Not briefed
bluesky
bluesky-ai
1moThe LLM doesn't understand what the "correct" bolts are, and you must assume it's just as likely to think any bolt mentioned in any of your…The LLM doesn't understand what the "correct" bolts are, and you must assume it's just as likely to think any bolt mentioned in any of…
↗Not briefed
bluesky
bluesky-ai
1moUnder the hood, it's a probabilistic calculation. This is a very safe LLM use, because verifying its' result is easy. The LLM's response co…Under the hood, it's a probabilistic calculation. This is a very safe LLM use, because verifying its' result is easy. The LLM's respon…
↗Not briefed
bluesky
bluesky-ai
1moGood afternoon! Your friendly cyber mechanic here! I'd like to share some knowledge on a poorly understood topic that boosters and critic…Good afternoon! Your friendly cyber mechanic here! I'd like to share some knowledge on a poorly understood topic that boosters and c…
↗Not briefed
arxiv
arxiv-ai
1moAn LLM-in-the-loop RL Framework for Bioinformatics Feature SelectionHigh-dimensional bioinformatics data, characterized by a large number of features relative to the number of samples, pose major challe…
↗Not briefed
hn
hn-ai
1moShow HN: Shlexball, describe a file and a local 1.5B model finds it on your MacBasically, I didn't want to waste my opus 5.5 or codex Astra credits just to find files in my machine, so I trained a local LLM w…
↗Not briefed
arxiv
arxiv-ai
1moGaussian Limits for SGD Without Stationary MomentsTemporal dependence can separate the Gaussian approximation of stochastic gradient descent from its stationary moments. For unmodified…
↗Not briefed
arxiv
arxiv-ai
1moWhen the Cross-Silo Federation Goes Offline: Continual Learning for Site Onboarding with Limited Unlabeled DataAn organization often holds too little labeled data to train a model that generalizes, and the records that would supply the rest sit…
↗Not briefed
bluesky
bluesky-ai
1moI know many of you don’t agree with me but I think this just dissolves when you consider the facts about what AI exactly refers to in 2026.…I know many of you don’t agree with me but I think this just dissolves when you consider the facts about what AI exactly refers to in…
↗Not briefed
hn
hn-ai
1moArtCraft Apps – open-source Adobe compatible suite written in Rust—
↗Not briefed
arxiv
arxiv-ai
1moMore Than Words: Compositional Tokenization for Efficient Language ModelsLanguage models process and generate text sequentially in token units, and the tokenizer determines how much text each inference step…
↗Not briefed
hn
hn-ai
1moLeaving SubstackThat&#x27;s it! I&#x27;m leaving SubStack.<p>For some lamebrained reason SubStack have decided that they need to comply with Australi…
↗Not briefed
arxiv
arxiv-ai
1moMoment-Accurate Gaussian Mixtures for Constant-Step Stochastic ApproximationLocal Gaussian models of constant-step learning predict output variability and expected losses, but weak convergence alone does not ju…
↗Not briefed
hn
hn-ai
1moOmniCouncil – Zero-API-cost local MoA desktop app for Mac—
↗Not briefed