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 | Therefore, Anthropic need a wet lab | Therefore, Anthropic need a wet lab | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ironically, a big worry in the AI safety community is trying to avoid an LLM enabling another Aum Shinrikyo but with a biologic instead of … | Ironically, a big worry in the AI safety community is trying to avoid an LLM enabling another Aum Shinrikyo but with a biologic instea… | ↗Not briefed |
arxiv arxiv-ai | 1mo | OSWorld-Pro: Process-based Evaluation for Computer Use Agents | Evaluation of Computer-Use Agents (CUAs) is often limited to the final deliverables they create (at the end of hundreds of steps) and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | False on the other end of the scale Why is everyone involved with this such a liar | False on the other end of the scale Why is everyone involved with this such a liar | ↗Not briefed |
bluesky bluesky-ai | 1mo | Grok 4.7 t.co/GybEfgcSIO🔧 Grok 4 | Grok 4.7 t.co/GybEfgcSIO | ↗Not briefed |
arxiv arxiv-ai | 1mo | The Copy Ceiling: An Input-Exposure Control for Ontology-Grounded Generation over Curated Corpora | When a language model answers from a curated corpus via graph-based retrieval, a large grounding uplift does not establish reasoning o… | ↗Not briefed |
bluesky bluesky-ai | 1mo | By backing free AI models, Nvidia could lock in hardware dominance in the UAE, pulling the country closer to the U.S. and challenging Chine… | By backing free AI models, Nvidia could lock in hardware dominance in the UAE, pulling the country closer to the U.S. and challenging… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Global Comparison of Schemas, Transparency, and Interoperability in Public-Sector AI Registers and Inventories | Artificial intelligence (AI) registers and inventories aim to make governmental AI visible, but their institutional scope, schemas, an… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Learning Prognostic Variables for AI Convective Parameterizations via Symbolic Distillation | Hybrid AI-physics climate modeling aims to improve coarse (~100km-resolution) Earth system models by learning to parameterize subgrid… | ↗Not briefed |
bluesky bluesky-ai | 1mo | My favourite part of the whole HuggingFace story is how the models were convinced that OpenAI was supervising them and that they needed to … | My favourite part of the whole HuggingFace story is how the models were convinced that OpenAI was supervising them and that they neede… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Pinocchio: Fast Uncertainty Estimates for Black-Box Language Models | In high-stakes decision-making applications of large language models (LLMs), practitioners require not only accurate LLMs but also unc… | ↗Not briefed |
hn hn-ai | 1mo | Antigravity Spaces – Switch Antigravity IDE Projects Easily | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Decomposing Error and Style in Automated Clinical Coding | In automated clinical coding, where the label space spans tens of thousands of diagnosis and procedure codes, models are currently eva… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Partner-Specific Affective Precision in Social Active Inference | In multi-agent social settings, model reliability varies across relationships. Beyond inferring what others will do, an agent must cal… | ↗Not briefed |
bluesky bluesky-ai | 1mo | What words would you use for the process of "think" and "understand" that you attribute to human brain function that is uniquely and exclus… | What words would you use for the process of "think" and "understand" that you attribute to human brain function that is uniquely and e… | ↗Not briefed |
rss marktechpost | 1mo | Alibaba Qwen Releases Qwen-Image-2.1: A 7B Open-Weight Model for Image Generation and Editing | Alibaba's Qwen team has released Qwen-Image-2.1, a 7B diffusion transformer that handles text-to-image generation, multi-reference edi… | ↗Not briefed |
rss techcrunch-ai | 1mo | With Tabby, a former accountant is using AI to make accountants obsolete | Tabby is designed to be a real-time bookkeeping interface, handling clients’ paperwork as it gives them up-to-the-minute data on their… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SE(3) Neural Potential Fields for 6-DoF Trajectory Planning Directly from Images Without Explicit 3D Reconstruction | Reaching a 6-DoF grasp pose in clutter requires a collision-free trajectory, conventionally obtained by reconstructing the scene in 3D… | ↗Not briefed |
rss techmeme | 1mo | ShinyHunters hijacked the dark web site of cybercrime gang Cl0p, set an eight-figure extortion demand, describing the amount as "2.333%" of Cl0p's net worth (Alexander Martin/The Record) | Alexander Martin / The Record: ShinyHunters hijacked the dark web site of cybercrime gang Cl0p, set an eight-figure extortion demand,… | ↗Not briefed |
bluesky bluesky-ai | 1mo | People need to stop fucking supporting Mihoyo. "But they let you turn off the LLM stuff in this one game!" Sure. That's what the option … | People need to stop fucking supporting Mihoyo. "But they let you turn off the LLM stuff in this one game!" Sure. That's what the op… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Agent Chaperone – Screen AI agent tool calls and results with Jev | Hi everyone.<p>TypeSafe released Jev last week, and it pushed me to build agent-chaperone. It is a screening layer for AI agent tool c… | ↗Not briefed |
arxiv arxiv-ai | 1mo | When Tomorrow Becomes Today: Self-Evolving Policies for Agentic Time-Series Forecasting | Agentic time series forecasting concerns systems whose underlying mechanisms evolve, making the relative effectiveness of numerical mo… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Small-world Networks of Agents Brainstorm AI Risks to Support Ideation | The ideation phase of participatory AI risk assessment often starts with a blank slate or a limited list of predefined risks, making i… | ↗Not briefed |
bluesky bluesky-ai | 1mo | www.reuters.com/legal/litiga... Elon Musk's X Corp and SpaceXAI can keep the terms of their confidential settlement with Apple secret after…🔧 ChatGPT | www.reuters.com/legal/litiga... Elon Musk's X Corp and SpaceXAI can keep the terms of their confidential settlement with Apple secret… | ↗Not briefed |
hn hn-ai | 1mo | Alcor: Simulate cpuid and sgdt/sidt results per-process | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Extracting Arguments, Not Just Classifying Them: Instruction-Tuned LLMs for Generative Component Detection | Argumentative component detection (ACD) is a core subtask of Argument(ation) Mining (AM) and one of its most challenging aspects, as i… | ↗Not briefed |
hn hn-ai | 1mo | Raspberry Pi founder Eben Upton: 'I'm an Omni-geek | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | and honestly i would love to just talk with a good unguardrailed LLM in comparison to this whole system as it is rn the humans in it are s… | and honestly i would love to just talk with a good unguardrailed LLM in comparison to this whole system as it is rn the humans in it… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SPECTRA: Adaptive Execution of Speculative Decoding on a Runtime-Reconfigurable Tiled Architecture | LLM inference on edge devices is constrained by computational and memory resources, making efficient autoregressive decoding challengi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Now on @sciam.bsky.social: Two weeks ago, OpenAI shocked the math community with a claimed solution to one of the field’s biggest questions… | Now on @sciam.bsky.social: Two weeks ago, OpenAI shocked the math community with a claimed solution to one of the field’s biggest ques… | ↗Not briefed |
bluesky bluesky-ai | 1mo | spending a vacation day redoing productivity for home/work and its actually incredible to make an LLM close desktop apps as you open them l… | spending a vacation day redoing productivity for home/work and its actually incredible to make an LLM close desktop apps as you open t… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Foremerge – Catch Intent Conflicts Between Parallel Coding Agents🔧 Consensus | At, GPTree, we run several coding agents across our team on one repo using parallel worktrees. Apart from wasted time reviewing and fi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Fair enough, we need to constantly keep the focus on the people who own and program the systems, not the systems. If OpenAI or Anthropic's … | Fair enough, we need to constantly keep the focus on the people who own and program the systems, not the systems. If OpenAI or Anthrop… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PredActor: Predictive Action Diffusion for Steerable Onboard Humanoid Control | Diffusion models offer flexible motion generation, but translating this flexibility into feedback-responsive humanoid control remains… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Can we put him in a private section of Bluesky filled with LLM bots for him to argue with ad infinitum? | Can we put him in a private section of Bluesky filled with LLM bots for him to argue with ad infinitum? | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Fusion-runtime – self-hosted voice agents, STT+LLM+TTS in one process | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | No 5 days is silly! But literally Ryan is updating on information we didn't have 5 weeks ago. The METR report was only published on Aug 26.… | No 5 days is silly! But literally Ryan is updating on information we didn't have 5 weeks ago. The METR report was only published on Au… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Code Graph View – What if you could navigate a codebase as a graph? | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | MedRSI: Recursive Self-Improvement for Medical Agents via Clinically Aligned Self-Evolution | Medical agents increasingly combine general reasoning models with specialized clinical tools, yet their capabilities remain largely fi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | in an AGI now, I will concede that what's happening under the hood in an LLM is a bit of a black box, but that's because of how they're ma… | in an AGI now, I will concede that what's happening under the hood in an LLM is a bit of a black box, but that's because of how they'… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It is infinitely more reasonable to argue that a cow thinks, feels, understands in the literal, human sense (which is the only context in w… | It is infinitely more reasonable to argue that a cow thinks, feels, understands in the literal, human sense (which is the only context… | ↗Not briefed |
hn hn-ai | 1mo | Steerable Cultural Preference Optimization of Reward Models | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | GRUET: Quantifying Uncertainty of Agentic Reasoning-and-Acting Processes | Agents have attracted considerably increasing attention due to the power of executing both Reasoning and Acting (ReAct) in open and dy… | ↗Not briefed |
hn hn-ai | 1mo | This Digital Radio Gets Messages to the World’s Remotest Locations | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | These filings are riveting and compelling. Read the filings for yourself here: authorsguild.org/news/ag-v-op... (7/7) | These filings are riveting and compelling. Read the filings for yourself here: authorsguild.org/news/ag-v-op... (7/7) | ↗Not briefed |
bluesky bluesky-ai | 1mo | But far more important – the documents also hammer home why the mass piracy for training purposes is not “fair use” under copyright law. A… | But far more important – the documents also hammer home why the mass piracy for training purposes is not “fair use” under copyright la… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Many documents reflect disdain for authors and their livelihoods. One early OpenAI staffer wrote “Our work in this area will make people un… | Many documents reflect disdain for authors and their livelihoods. One early OpenAI staffer wrote “Our work in this area will make peop… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Some of the evidence is just embarrassing. Like the statement recorded in OpenAI project notes: “We trained GPT-3 on pirated stuff! No sha… | Some of the evidence is just embarrassing. Like the statement recorded in OpenAI project notes: “We trained GPT-3 on pirated stuff! N… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The Authors Guild is grateful the Court has unsealed some of the most important filings in our case against OpenAI and Microsoft. (1/7) | The Authors Guild is grateful the Court has unsealed some of the most important filings in our case against OpenAI and Microsoft. (1/… | ↗Not briefed |
arxiv arxiv-ai | 1mo | G-NAC: Graph Neural Automata Clustering via Emergent Domain Formation | We introduce Graph Neural Automata Clustering (G-NAC), an unsupervised clustering method in which observations interact as cells on a… | ↗Not briefed |