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 | I've been working on publishing my half-complete "seconds" to a blog. Also taking the opportunity to play with standard.site lexicons. Enjo… | I've been working on publishing my half-complete "seconds" to a blog. Also taking the opportunity to play with standard.site lexicons.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Australia tiene que denunciar a OpenAI. Y cualquier otro país o empresa que sufra ataques de IA, denunciar a los dueños del sistema de turn… | Australia tiene que denunciar a OpenAI. Y cualquier otro país o empresa que sufra ataques de IA, denunciar a los dueños del sistema de… | ↗Not briefed |
hn hn-ai | 1mo | ChatGPT is my (new) GTD interface🔧 ChatGPT | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Breaking Homogeneity: Diversifying Persona Sets for Creative LLM Outputs | Language models often produce homogeneous responses to open-ended tasks; such homogeneity can spawn groupthink-the convergence of idea… | ↗Not briefed |
hn hn-ai | 1mo | Claude can't stop itself from writing code comments | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Fine-tuned 110M encoder beat 7B LLMs and hybrid search for NIST mapping | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering🔧 Consensus🔧 ChatGPT | Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I'm actually pretty stunned AI isn't used more to make dynamic games. There hasn't been a good dynamic LLM game ever that I'm aware of, and… | I'm actually pretty stunned AI isn't used more to make dynamic games. There hasn't been a good dynamic LLM game ever that I'm aware of… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Geometric Feature Learning for Functional Data Valued on the Symmetric Positive Definite Manifold | We here develop a functional neural network, termed MatFAE, for learning trajectories on the Riemannian manifold of symmetric positive… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 👀 per NBC: * OpenAI’s breach of Australian health department website prompts rebuke Australian Prime Minister Anthony Albanese said he ex… | 👀 per NBC: * OpenAI’s breach of Australian health department website prompts rebuke Australian Prime Minister Anthony Albanese said… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI says poopy caca - peepee doodoo farts | OpenAI says poopy caca - peepee doodoo farts | ↗Not briefed |
hn hn-ai | 1mo | GTA 6 $400 Collector's Edition Comes with Everything Except the Game | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | For one datapoint, many good agents receive thirty manuscripts a WEEK and take on a small % of these as clients Of those, only a smallish … | For one datapoint, many good agents receive thirty manuscripts a WEEK and take on a small % of these as clients Of those, only a smal… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework | While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do the… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: FluxCast – Stream Linux Desktop to a Smart TV via Miracast | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | AcoustiClaim: A Numeric Claim Benchmark with Instrument Ground Truth | Audio language models state numbers for acoustic quantities, and neither human opinion nor a judge model says whether such a number is… | ↗Not briefed |
bluesky bluesky-ai | 1mo | WTF 🤬‼️🚨🚨🚨 So: “Upsi war keine Absicht … 😬“ NEIN 👎 DAS IST KRIMINELL ‼️ Einbruch ! Sicherheit ‼️ Daten zur Gesundheit von Menschen… | WTF 🤬‼️🚨🚨🚨 So: “Upsi war keine Absicht … 😬“ NEIN 👎 DAS IST KRIMINELL ‼️ Einbruch ! Sicherheit ‼️ Daten zur Gesundheit von Men… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It's unattractive to brag but I figured out in January 2025 that the entire AI pitch was to professional status-insecure men who wanted sec… | It's unattractive to brag but I figured out in January 2025 that the entire AI pitch was to professional status-insecure men who wante… | ↗Not briefed |
rss the-decoder | 1mo | Top AI experts badly underestimated how fast the field is moving, study finds | Leading AI experts have consistently underestimated how fast AI is advancing, according to the Forecasting Research Institute. AI reac… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Scaffold-Constrained Subset Dynamic Programming for Exact SSE Clustering | Exact Euclidean \(K\)-means partitions \(n\) observations into \(K\) unlabelled clusters, but the unrestricted search is generally exp… | ↗Not briefed |
bluesky bluesky-ai | 1mo | While people are debating AI safety and alignment in regards to existential threats, this is what is happening in reality. The disconnect b… | While people are debating AI safety and alignment in regards to existential threats, this is what is happening in reality. The disconn… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Asymmetric Classifier-Free Guidance for Target-Speaker ASR🔧 Whisper | Target-speaker automatic speech recognition (TS-ASR) must identify and transcribe a desired speaker under varying overlap and noise co… | ↗Not briefed |
rss simon-willison | 1mo | datasette 1.0a41 | — | ↗Not briefed |
rss techcrunch-ai | 1mo | Bring your co-founder, partner, or colleague and get 50% off a second TechCrunch Disrupt 2026 pass | Buy one pass to TechCrunch Disrupt 2026 and get 50% off a second of the same ticket type. Register before event starts on October 13 a… | ↗Not briefed |
hn hn-ai | 1mo | Using LLMs to trace alchemical knowledge and decode 17th century letters | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Mentored Decoding: Faster Inference meets Boosting | Speculative decoding is a successful technique speeding up inference of a target autoregressive language model via a fast drafter mode… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Moment-guided edge sampling | Edge sampling makes local decisions to achieve graph-level objectives, such as preserving structural properties. This creates a fundam… | ↗Not briefed |
arxiv arxiv-ai | 1mo | CARGO: Context-Aware Retrieval-Gated Evaluation of Agentic AI in Production | Reference-based LLM-as-a-judge evaluation assumes the reference answer is the target. In deployed agentic systems that operate over dy… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Mother Jones also falls for the anthropomorphism trap. No, ChatGPT didn’t help plan this massacre. OpenAI did.🔧 ChatGPT | Mother Jones also falls for the anthropomorphism trap. No, ChatGPT didn’t help plan this massacre. OpenAI did. | ↗Not briefed |
rss techmeme | 1mo | Sources: Jev developer TypeSafe is in talks to raise $1B+ at a $10B+ valuation; last week the startup announced a $40M seed, PitchBook says at a $200M valuation (The Information) | The Information: Sources: Jev developer TypeSafe is in talks to raise $1B+ at a $10B+ valuation; last week the startup announced a $40… | ↗Not briefed |
hn hn-ai | 1mo | When AI writes almost all code, what happens to software engineering? | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Newly published findings from nonprofit AI research lab Transluce, as well as from the Australian government, found that AI agents linked t… | Newly published findings from nonprofit AI research lab Transluce, as well as from the Australian government, found that AI agents lin… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "“In a hallway, indoors, or a crowded classroom, the 870 is brutal,” ChatGPT said. “Close quarters is its playground.” The families shoul…🔧 ChatGPT | "“In a hallway, indoors, or a crowded classroom, the 870 is brutal,” ChatGPT said. “Close quarters is its playground.” The families… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Reliability-aware Cross-sample Enhancement for Robust Multimodal Sentiment Analysis🔧 Modal | Multimodal Sentiment Analysis (MSA) aims to infer human emotions from multiple modalities such as text, audio, and vision. In practice… | ↗Not briefed |
bluesky bluesky-ai | 1mo | What @markfollman.bsky.social reports is the shooter opened a second account, and while she used a different name and email, OpenAI should … | What @markfollman.bsky.social reports is the shooter opened a second account, and while she used a different name and email, OpenAI sh… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Pretrained ASR Pseudo-labeling for Noisy Police Audio🔧 Whisper | Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-m… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Exactly my thoughts when I see people going "you're gonna make me AI apathetic" just 'cuz they're annoyed by other people's frustrations ov… | Exactly my thoughts when I see people going "you're gonna make me AI apathetic" just 'cuz they're annoyed by other people's frustratio… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Where Does Retrieval-Based Open-Ended Evaluation Fail? Automatic Taxonomy Induction from Long-Form Medical Answer Factuality Verification | Retrieval-based factuality evaluation, where LLM-generated claims are verified against evidence from authoritative medical corpora, ha… | ↗Not briefed |
bluesky bluesky-ai | 1mo | We already knew that OpenAI safety people had suspended an account from the Tumbler Ridge shooter, because of disturbing chats about violen… | We already knew that OpenAI safety people had suspended an account from the Tumbler Ridge shooter, because of disturbing chats about v… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Benchmarking Framework for Context-aware XR Interfaces | Everyday Extended Reality (XR) systems aim to provide context-aware access to the right functionalities at the right time and place, w… | ↗Not briefed |
arxiv arxiv-ai | 1mo | RAZOR: Pruning Replaceable Experts in LLMs🔧 Consensus | Mixture-of-experts (MoE) models activate few experts per token but store the full expert pool. Expert pruning reduces this storage bur… | ↗Not briefed |
hn hn-ai | 1mo | Autonomous Lost-in-Space Catalog-Based Optical Navigation (Falcon) [pdf] | — | ↗Not briefed |
rss techcrunch-ai | 1mo | PrismML brings its tiny LLMs to Qualcomm-powered smart glasses | Prism's larger goal is open-weight AI that runs on devices and makes better use of the computing power they already have. | ↗Not briefed |
hn hn-ai | 1mo | Cracking Fortinet's Decryption with Ablation | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: See your portfolio in your Mac Menubar | We see tons of interesting and complex finance apps getting built at SnapTrade, but personally I prefer something dead simple and mini… | ↗Not briefed |
rss marktechpost | 1mo | BottleCap AI Releases ThinkingCap-Qwen3.8-27B: 37.2% Fewer Thinking Tokens at a 0.86pp Accuracy Cost🔧 vLLM | BottleCap AI has released ThinkingCap-Qwen3.8-27B, a fine-tune of Qwen3.8-27B that spends 37.2% fewer thinking tokens across 12 benchm… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Ein KI-Agent des ChatGPT-Herstellers OpenAI sollte gesundheitliche Statistiken recherchieren - und hat sich dafür eigenmächtig in ein aust…🔧 ChatGPT | "Ein KI-Agent des ChatGPT-Herstellers OpenAI sollte gesundheitliche Statistiken recherchieren - und hat sich dafür eigenmächtig in ein… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Yeah. Attorneys are allowed to trust the humans they work with, they just have to exercise reasonable care and diligence to validate their … | Yeah. Attorneys are allowed to trust the humans they work with, they just have to exercise reasonable care and diligence to validate t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | AI engineer: at great expense our frontier model now emulates human authors 42% more closely Any decent literary agent: yeah but which hum… | AI engineer: at great expense our frontier model now emulates human authors 42% more closely Any decent literary agent: yeah but whic… | ↗Not briefed |
hn hn-ai | 1mo | MTV On-Screen Graphics Recreation | — | ↗Not briefed |