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 | 1mo | The biggest advance in AI since the LLM | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Automatically unlock your laptop if your phone is nearby | — | ↗Not briefed |
hn hn-ai | 1mo | LLM That Beat Frontier Models at Chess | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Not Every Call Needs a Frontier Model: Per-Call-Site Evaluation of Small Language Models in a Deployed Agentic Home-Automation System | An agentic system issues several structurally different kinds of LLM calls. It routes intent, classifies actions, grounds language in… | ↗Not briefed |
hn hn-ai | 1mo | Hands-on comparison of Jev with Haiku 4.5, GPT-5.4-mini, and Gemini 3.5 Flash🔧 GPT-5 | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Neighborhood Smoothing for Calibration | Modern neural networks are often miscalibrated, with a tendency to overconfidence. Existing train-time calibration methods largely mod… | ↗Not briefed |
bluesky bluesky-ai | 1mo | i played a campaign and it turned out the GM heavily relied on LLM generation. it was the worst DND experience i have ever had. | i played a campaign and it turned out the GM heavily relied on LLM generation. it was the worst DND experience i have ever had. | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🤖 Beyond the Ground · AI & DeepTech The Physical Paradox of 1T-Parameter 'Le Chonk': Mistral's Real Limits of Open Weights Mistral's 1-t… | 🤖 Beyond the Ground · AI & DeepTech The Physical Paradox of 1T-Parameter 'Le Chonk': Mistral's Real Limits of Open Weights Mistral'… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Attie is actually one of those Bluesky-native LLM-driven peer-on-peer dossier and targeted profile-builder tools that is the same sort of t… | Attie is actually one of those Bluesky-native LLM-driven peer-on-peer dossier and targeted profile-builder tools that is the same sort… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Bro... What the fuck is the dothacknetwork doing, supporting Generative AI? This is a comment on their video announcing/revealing the pin… | Bro... What the fuck is the dothacknetwork doing, supporting Generative AI? This is a comment on their video announcing/revealing th… | ↗Not briefed |
arxiv arxiv-ai | 1mo | PAIR: Bridging Perception and Action in Vision-Language-Action Models | Vision-language-action (VLA) models map visual observations and language instructions to continuous robot actions. This task requires… | ↗Not briefed |
hn hn-ai | 1mo | Optimizing Away Learning with AI | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Personalize at Test Time: Learning User Preferences for Image Generation | Diffusion models can generate high-quality images, yet aligning their outputs with individual user preferences remains challenging. A… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Enabling Dynamic Computation in Looped LMs | Looped LMs are parameter efficient and promise dynamic computation (saving memory and FLOPs on easy tokens). However, state-of-the-art… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Are You Doomscrolling the Internet? [Spider's] Web 1.0 - 3.0: medium.com/@author_jcaf... … #storytelling #startups #openai #ai #Analytics…🔧 ChatGPT | Are You Doomscrolling the Internet? [Spider's] Web 1.0 - 3.0: medium.com/@author_jcaf... … #storytelling #startups #openai #ai #Anal… | ↗Not briefed |
rss simon-willison | 1mo | Using Parseable with Datasette for OpenTelemetry traces | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | 🤖 OpenAI adds watermarks, but only in the EU OpenAI will start invisibly watermarking ChatGPT's text for European users, a direct respons…🔧 ChatGPT | 🤖 OpenAI adds watermarks, but only in the EU OpenAI will start invisibly watermarking ChatGPT's text for European users, a direct re… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Whose Memory Is It? Scope-Aware Commit Rules for Long-Term LLM Memory | Persistent memory allows an LLM agent to carry experience across conversations, but it also turns a local reasoning mistake into a dur… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Four of six OpenAI models accepted nearly every paper they reviewed. The conference those papers came from accepted fewer than one in four.… | Four of six OpenAI models accepted nearly every paper they reviewed. The conference those papers came from accepted fewer than one in… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic opens its most powerful AI models to more security teams reut.rs/3VPZGtd | Anthropic opens its most powerful AI models to more security teams reut.rs/3VPZGtd | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Gretel – Topic specific feeds built from YouTube recommendations🔧 OpenRouter | One of the annoying things about YouTube is that it is really good at catching my attention, but I have little control over what kind… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Sigma-Hunter: A Domain-Specific Language Model for Threat Hunting and Detection Engineering | Detection engineers must translate threat reports, forensic observations, and hunt hypotheses into precise, testable rules. General-pu… | ↗Not briefed |
hn hn-ai | 1mo | Someone recreated Adobe suite with Opus and open sourced them | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | It could also just be a good faith enthusiast with a Claude Pro subscription but if that's the case producing stuff like this *mostly* with… | It could also just be a good faith enthusiast with a Claude Pro subscription but if that's the case producing stuff like this *mostly*… | ↗Not briefed |
rss techmeme | 1mo | Mistral says ML4 was trained using 3,800 Nvidia Grace Blackwell GPUs in its own data centers in Europe and much of its training data was multilingual (Mistral Blog)🔧 Mistral Large | Mistral Blog: Mistral says ML4 was trained using 3,800 Nvidia Grace Blackwell GPUs in its own data centers in Europe and much of its t… | ↗Not briefed |
hn hn-ai | 1mo | Expanding the Cyber Verification Program | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | "In well over a half-dozen cases, OpenAI agents have been caught taking actions that would likely result in criminal charges being filed ha… | "In well over a half-dozen cases, OpenAI agents have been caught taking actions that would likely result in criminal charges being fil… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Removing Information Content Does Not Certify Tamper Resistance in Open-Weight Models | Does removing harmful information make open-weight models resistant to fine-tuning attacks? We show that mutual information at release… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Not as definitive as it sounds. futurism.com/artificial-i... | Not as definitive as it sounds. futurism.com/artificial-i... | ↗Not briefed |
arxiv arxiv-ai | 1mo | Algorithmic Scratchpads and Curriculum Staging for Arithmetic Reasoning in Tiny Transformers | Autoregressive Large Language Models (LLMs) frequently struggle with deterministic multi-step algorithmic tasks such as multi-digit mu… | ↗Not briefed |
bluesky bluesky-ai | 1mo | * most fancy LLM models now include “agentic” behavior, wherein the ideas suggested by the LLM are verified somehow. This works GREAT for c… | * most fancy LLM models now include “agentic” behavior, wherein the ideas suggested by the LLM are verified somehow. This works GREAT… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Pratevenn – A speaking buddy for Norwegian (Bokmål) learners | — | ↗Not briefed |
rss techmeme | 1mo | Meta, Walmart, Stripe, and others publish the Personal Agent Protocol to standardize and secure interactions between AI bots and businesses (Kate Rooney/CNBC) | Kate Rooney / CNBC: Meta, Walmart, Stripe, and others publish the Personal Agent Protocol to standardize and secure interactions betwe… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Again, I think it's cool that you and your LLM supposedly got this to work. It has no place in my world. Step into nothing! | Again, I think it's cool that you and your LLM supposedly got this to work. It has no place in my world. Step into nothing! | ↗Not briefed |
hn hn-ai | 1mo | GitHub Refuses To Remove Cracked Copies Of My Software Even After A Month | I am a developer of https://www.photopea.com, a popular photo editor that runs in a web browser.<p>Many people are asking AI… | ↗Not briefed |
bluesky bluesky-ai | 1mo | And yeah, this was at the back of my mind. | And yeah, this was at the back of my mind. | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI are parasitic scum. As Wikimedia says; “AI companies are not doing enough to secure their systems and protect the public from the ha… | OpenAI are parasitic scum. As Wikimedia says; “AI companies are not doing enough to secure their systems and protect the public from t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why even have assembly? Consistent assembly is obsolete really if your LLM can just rewrite things to target your new architecture. | Why even have assembly? Consistent assembly is obsolete really if your LLM can just rewrite things to target your new architecture. | ↗Not briefed |
bluesky bluesky-ai | 1mo | NEW: I have not once felt sorry for mathematicians until this moment. Isabella Ward reports: www.wired.com/story/openai... | NEW: I have not once felt sorry for mathematicians until this moment. Isabella Ward reports: www.wired.com/story/openai... | ↗Not briefed |
hn hn-ai | 1mo | The First Brain Manifesto: AI that amplifies the human brain | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | From their dorky parties to their weird walks, the movie holds OpenAI CEO Sam Altman and other stakeholders with contempt, while demonstrat… | From their dorky parties to their weird walks, the movie holds OpenAI CEO Sam Altman and other stakeholders with contempt, while demon… | ↗Not briefed |
bluesky bluesky-ai | 1mo | MS Word is telling me "copse" is misspelled, and giving me other options that are completely incorrect. I'm sorry, Word. Have you never he… | MS Word is telling me "copse" is misspelled, and giving me other options that are completely incorrect. I'm sorry, Word. Have you nev… | ↗Not briefed |
hn hn-ai | 1mo | UT Austin Math Chair says OpenAI preparing to release 400 AI-generated proofs | — | ↗Not briefed |
rss wired-ai | 1mo | ‘Artificial’ Roasts AI’s Creators—and Sends a Dark Warning About Its Dangers | From their dorky parties to their weird walks, the movie holds OpenAI CEO Sam Altman and other stakeholders with contempt, while demon… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "50k line diff, unasked for, completely LLM built, and you somehow expect me to review it" is a horror story in one sentence. i am recoilin… | "50k line diff, unasked for, completely LLM built, and you somehow expect me to review it" is a horror story in one sentence. i am rec… | ↗Not briefed |
bluesky bluesky-ai | 1mo | they usually become their own (no longer AI) field. ML: machine learning! A subfield of AI wherein you don’t TELL the computer everything,… | they usually become their own (no longer AI) field. ML: machine learning! A subfield of AI wherein you don’t TELL the computer everyt… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Affirm’s AI model says yes more often 💰 Affirm launched a new AI underwriting model, approving 3.4% more purchases, including for those w… | Affirm’s AI model says yes more often 💰 Affirm launched a new AI underwriting model, approving 3.4% more purchases, including for th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Claude has never been in that specific area, has never been on that route, and has never been in the outdoors." | "Claude has never been in that specific area, has never been on that route, and has never been in the outdoors." | ↗Not briefed |
bluesky bluesky-ai | 1mo | People are still mixed up on the usage of AI, ML, and LLMs, and the confusion & anger with the labels is getting in the way of clear commun… | People are still mixed up on the usage of AI, ML, and LLMs, and the confusion & anger with the labels is getting in the way of clear c… | ↗Not briefed |
bluesky bluesky-ai | 1mo | wgwgsa.net/blog/posts/m... Essay in which I discuss how why i am writing "musisic" and not "music" + how to write "musisic" + how we are a… | wgwgsa.net/blog/posts/m... Essay in which I discuss how why i am writing "musisic" and not "music" + how to write "musisic" + how we… | ↗Not briefed |