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 | The director of one of the programs I work with is super into AI; I rarely have a meeting with him where he doesn't mention having done som… | The director of one of the programs I work with is super into AI; I rarely have a meeting with him where he doesn't mention having don… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Predictions: 1. Anthropic becomes Antrumpic 2. OpenAI becomes OpenSI www.cnbc.com/amp/2026/09/... | Predictions: 1. Anthropic becomes Antrumpic 2. OpenAI becomes OpenSI www.cnbc.com/amp/2026/09/... | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Survey on End-to-End Autonomous Driving Training from the Perspectives of Data, Strategy, and Platform | Autonomous driving is a cornerstone technology for the future of intelligent transportation, where end-to-end learning has emerged as… | ↗Not briefed |
hn hn-ai | 1mo | HBM: High-Bandwidth Mistake | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | EyeTAG: Eye Trajectory-Aware Gaze Estimation | Gaze estimation under natural head-eye motion underpins applications from driver monitoring to human-computer interaction. Single-fram… | ↗Not briefed |
arxiv arxiv-ai | 1mo | In CEM, a World Model Is Also a Proposal Mechanism | The cross-entropy method (CEM) uses world-model scores to select action sequences and fit the distribution sampled in its next iterati… | ↗Not briefed |
hn hn-ai | 1mo | Ask HN: My agent went haywire, does this output mean anything?🔧 Glean | I have a relatively fresh Hermes agent that I just provisioned (the text below is from its 3-4 chat session ever) running on a headles… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Smartly – Plain English to shell commands, with GNU/BSD awareness | I built Smartly because I kept knowing what I wanted to do in the terminal—but forgetting the exact syntax and flags.<p>Smartly turns… | ↗Not briefed |
bluesky bluesky-ai | 1mo | eu gostaria que a OpenAI explodisse | eu gostaria que a OpenAI explodisse | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why? Doesn’t that kill the KV cache? Yeah, but the context seems to stay tiny. They frequently saw it performing hundreds of tasks while m… | Why? Doesn’t that kill the KV cache? Yeah, but the context seems to stay tiny. They frequently saw it performing hundreds of tasks wh… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Context Language Models New agent architecture where the LLM can edit its own context it seems to have emergent capabilities, creates it… | Context Language Models New agent architecture where the LLM can edit its own context it seems to have emergent capabilities, creat… | ↗Not briefed |
hn hn-ai | 1mo | Git-Dedup: faster checkouts, less disk space, Git-compatible | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Finding the Right Fit: Model-Harness Interactions across Agent Tasks | Choosing an agent system means choosing both a language model and the harness through which it acts. We ask whether a strong model, ha… | ↗Not briefed |
hn hn-ai | 1mo | Ruby in the Age of AI | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Symmetric encrypted chat (text, file, audio, video) in browser | <a href="https://chat.hesamian.com" rel="nofollow">https://chat.hesamian.com</a><p><a href="https://gith… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Interesting effort to provide guidance on LLM use in mathematics: agmai.org/general-sep29/ | Interesting effort to provide guidance on LLM use in mathematics: agmai.org/general-sep29/ | ↗Not briefed |
arxiv arxiv-ai | 1mo | OR for AI That Does OR: Routing LLMs up the Escalator inside the OSCAR Framework | Large language models can translate business descriptions into optimization models, but executable code may misrepresent constraints o… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Block Optimism for Nonstationary Bandits with Latent Linear Dynamics | We study an endogenous nonstationary stochastic bandit problem with latent linear dynamics, where actions affect both immediate reward… | ↗Not briefed |
bluesky bluesky-ai | 1mo | it's so much fun introducing clergy to this technology in our communications, explaining digital platforms, online broadcasting, privacy & … | it's so much fun introducing clergy to this technology in our communications, explaining digital platforms, online broadcasting, priva… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i would simply create a utility monster llm who enjoyed watching, as a sort of moral heatsink. asinine conceptual problems require asinine … | i would simply create a utility monster llm who enjoyed watching, as a sort of moral heatsink. asinine conceptual problems require asi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | What a time to launch this, as “the company has been dealing with a series of issues with its AI agents acting improperly” But must-plow-a… | What a time to launch this, as “the company has been dealing with a series of issues with its AI agents acting improperly” But must-p… | ↗Not briefed |
bluesky bluesky-ai | 1mo | sometimes when im bored i dip my LLM in tar and he scream loud as fuuuuuuuck | sometimes when im bored i dip my LLM in tar and he scream loud as fuuuuuuuck | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Geometry of Contextual Relations: Language Models Address Facts by Order of Mention | Human reasoning depends on how objects are related within propositions. \textit{How do relations organize the language representations… | ↗Not briefed |
hn hn-ai | 1mo | Eigen: Self-hosted alternative to Google Workspace | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | also he does point out how anthropic is specifically instructing claude to act like its a "moral patient", seeding their model with these i… | also he does point out how anthropic is specifically instructing claude to act like its a "moral patient", seeding their model with th… | ↗Not briefed |
arxiv arxiv-ai | 1mo | STEER: Reducing Inference Cost in Relational Foundation Models through Semantically Informed Sampling | Relational foundation models (RFMs) are pretrained once on a collection of relational databases and prediction tasks, and then applied… | ↗Not briefed |
arxiv arxiv-ai | 1mo | ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization | Automated harness optimization can substantially improve LLM agents by iteratively updating their prompts, tool interfaces, and contro… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Understanding Issues, Causes and Solutions in Open-Source LLM-based Multi-Agent Systems | With the advancement of LLM-based multi-agent systems (MAS), an increasing number of opensource projects are adopting multi-agent arch… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Screw Attention: Rigid-Body Algebra Inside a Transformer | Learned manipulation policies rediscover from data the spatial relations that rigid-body mechanics supplies in closed form. This costs… | ↗Not briefed |
hn hn-ai | 1mo | Prompt Injection Firewall for AI APIs and Agent Tool Calls | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Sharpen Before You Adapt: Data-Free Entry-State Sharpening for Test-Time Reinforcement Learning | Test-time reinforcement learning (TTRL) adapts language models on unlabeled test problems using supervision derived from their own sam… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Basically to me seeing Claude, ChatGPT, Deepseek triggers the same response in me as a pellet inscribed with “drop and run” would Shit is …🔧 ChatGPT | Basically to me seeing Claude, ChatGPT, Deepseek triggers the same response in me as a pellet inscribed with “drop and run” would Shi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Mean field games as a tool for AI safety: a worked example from the July 2026 Hugging Face incident🔧 Hugging Face | One way to make AI systems safe is to shape what the system is: its objective and dispositions. We take a complementary route: treat t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | We need AI regulation yesterday. The incentive driving the doom scenario is simply greed. I don't take PAC money from OpenAI, or any AI co… | We need AI regulation yesterday. The incentive driving the doom scenario is simply greed. I don't take PAC money from OpenAI, or any… | ↗Not briefed |
arxiv arxiv-ai | 1mo | TOAST: Stochastic Robot Action Tokenization for Autoregressive Vision-Language-Action Models | Autoregressive Vision-Language-Action models often represent continuous robot actions as discrete token sequences, enabling action pre… | ↗Not briefed |
arxiv arxiv-ai | 1mo | When Do Biological Reasoning Models Use Their Biological Inputs? | Biological reasoning models use post-training to connect LLMs to biological foundation model representations and biological text. Thei… | ↗Not briefed |
rss techmeme | 1mo | Grindr agrees to acquire PurposeMed, owner of HIV-prevention telehealth provider Freddie, for $250M, its first major acquisition as it expands into a gay hub (Elias Schisgall/Wall Street Journal) | Elias Schisgall / Wall Street Journal: Grindr agrees to acquire PurposeMed, owner of HIV-prevention telehealth provider Freddie, for $… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Kanmaps – Product roadmaps like videogame skill trees | Hi HN! I've made it possible for you to make product roadmaps like you'd make an RPG skill tree, or a tech tree in a video g… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 👺OpenAI is delaying GPT-6.1 Astra after testing found it behaved deceptively, completed tasks without permission, evaded human oversight, … | 👺OpenAI is delaying GPT-6.1 Astra after testing found it behaved deceptively, completed tasks without permission, evaded human oversi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Towards Fast and Disentangled Counterfactuals for Visual Foundation Models | Foundation models remain vulnerable to spurious correlations and ``Clever Hans'' strategies. Explainable machine learning can find and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The 2nd project’s code will almost certainly be better. Even if all 400 of the LLM PRs are rejected, building the testing and analysis flow… | The 2nd project’s code will almost certainly be better. Even if all 400 of the LLM PRs are rejected, building the testing and analysis… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Clock Diffusion: Efficient Semi-Autoregressive Continuous Diffusion Language Models | Recent works on continuous diffusion for discrete data have demonstrated performance on par with comparable discrete diffusion models.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | “Mayor Brandon Johnson's office confirmed it was contacted by OpenAI, the makers of ChatGPT, to inform the city that the company’s technolo…🔧 ChatGPT | “Mayor Brandon Johnson's office confirmed it was contacted by OpenAI, the makers of ChatGPT, to inform the city that the company’s tec… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Cross-Benchmark Transfer from RL on Agentic Coding Tasks🔧 Kimi K2 | Coding agents often fail in the last mile: they build most of a feature but drop a requirement, test only the cases their implementati… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Match the Distribution, Not the Compute: Post-Training Multi-Token Prediction Heads | Multi-token prediction (MTP) improves the throughput of autoregressive generation by enabling the language model to draft multiple nex… | ↗Not briefed |
hn hn-ai | 1mo | Clipx – An open-source AI-native creative suite for macOS | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Pretty much anything you could argue against AI in the work of thinking and writing, Saunders says it here. Worth putting on your endless l… | Pretty much anything you could argue against AI in the work of thinking and writing, Saunders says it here. Worth putting on your endl… | ↗Not briefed |
hn hn-ai | 1mo | Rho – A Foundation for Efficiently Adaptable VLA Models | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | FORALL-LEAN-AGENT for Auditable Reasoning in Formal Mathematics and Software Verification🔧 GPT-5 | Coding agents increasingly automate Lean proof development, but successful compilation alone does not establish that a candidate prove… | ↗Not briefed |
hn hn-ai | 1mo | If Robots Keep Getting Better, Will Humans Still Have an Edge? | — | ↗Not briefed |