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 |
|---|---|---|---|---|
arxiv arxiv-ai | 1mo | Right Choice of Classification Algorithms Based on Reinforcement Learning for Prediction of Non-Alcoholic Fatty Liver | There are many complex issues in the world of artificial intelligence. Some of these problems are solved using other artificial intell… | ↗Not briefed |
bluesky bluesky-ai | 1mo | BBC almost got there. I know it’s hard to do critical thinking but was ‘OpenAI hacks Australian Government’ just too difficult to write? … | BBC almost got there. I know it’s hard to do critical thinking but was ‘OpenAI hacks Australian Government’ just too difficult to wr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | ...either harming an ongoing life - by mutilating the homeostasis of a system which evolved a particular way for a reason - or having creat… | ...either harming an ongoing life - by mutilating the homeostasis of a system which evolved a particular way for a reason - or having… | ↗Not briefed |
bluesky bluesky-ai | 1mo | When is OpenAI going to face criminal prosecution for all the cybercrimes they've admitted to? | When is OpenAI going to face criminal prosecution for all the cybercrimes they've admitted to? | ↗Not briefed |
arxiv arxiv-ai | 1mo | Towards Deployable Underwater Vessel Classification | We propose a compact underwater acoustic classification framework combining multi-representation feature engineering, temporal statist… | ↗Not briefed |
bluesky bluesky-ai | 1mo | www.bbc.co.uk/news/live/cv... | www.bbc.co.uk/news/live/cv... | ↗Not briefed |
bluesky bluesky-ai | 1mo | So an AI system hacks into deeply private data and the company decides not to tell the government for months. Irresponsible doesn't begin t… | So an AI system hacks into deeply private data and the company decides not to tell the government for months. Irresponsible doesn't be… | ↗Not briefed |
bluesky bluesky-ai | 1mo | その『ゲーム離れ』に関しても、LLMが元凶であるって事実からは全力で逃げて「一体なぜ？🤔」とすっとぼけていくんだろうなって…んで何か適当に見当違いな考察()を幾つか並べて有耶無耶にするんだろうなって… | その『ゲーム離れ』に関しても、LLMが元凶であるって事実からは全力で逃げて「一体なぜ？🤔」とすっとぼけていくんだろうなって…んで何か適当に見当違いな考察()を幾つか並べて有耶無耶にするんだろうなって… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I have LOST MY SHIT about the OpenAI breach of Australia’s Medicare records in the Guardian: content warning, very angry. www.theguardian.… | I have LOST MY SHIT about the OpenAI breach of Australia’s Medicare records in the Guardian: content warning, very angry. www.theguar… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Once again, this is not an AI agent 'going rogue', this is OpenAI not doing their jobs properly. It is human error/ineptitude, not super-sm… | Once again, this is not an AI agent 'going rogue', this is OpenAI not doing their jobs properly. It is human error/ineptitude, not sup… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Spot, Separate, and Enhance: Fully Generative Approach for Audio Mixing | We introduce Spot, Separate, and Enhance (SSE), the first multimodal, user-guided generative model for audio remixing and enhancement.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Everything comes back to tech company liability. The DMCA and its equivalents meant tech companies/exexs took no responsibility, and were a… | Everything comes back to tech company liability. The DMCA and its equivalents meant tech companies/exexs took no responsibility, and w… | ↗Not briefed |
arxiv arxiv-ai | 1mo | IndicBankBench: Evaluating Safety and Reliability of Language Model Assistants in Indian Retail Banking | Banking assistants must use account-specific information to answer requests and, in many cases, take actions through tools. Evaluating… | ↗Not briefed |
arxiv arxiv-ai | 1mo | HarnessPAI: An Evolving Harness for Physical AI | Physical AI aims to build embodied agents that perceive the world, understand and reason about it, and decide how to act. Yet the fiel… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It is a huge shame to see so much language attributing agency, intent and thought to a software program, when all that should be attributed… | It is a huge shame to see so much language attributing agency, intent and thought to a software program, when all that should be attri… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The only LLM that I'LL ever fuck with is LONG LONG MAN | The only LLM that I'LL ever fuck with is LONG LONG MAN | ↗Not briefed |
arxiv arxiv-ai | 1mo | Edge AI on Constrained Devices for Binary Sleep-Wake Classification in Dynamic Environments | This paper presents an Edge AI-based system for detecting sleep and wake states in non-stationary mobile environments using resource-c… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Seriously beginning to wonder if this is an 'opportunity', either contrived or grasped, to ramp up the push for 'bigger n better' data cent… | Seriously beginning to wonder if this is an 'opportunity', either contrived or grasped, to ramp up the push for 'bigger n better' data… | ↗Not briefed |
bluesky bluesky-ai | 1mo | J'ai vu les captures d'écran d'un (long) thread twitter expliquant que des métaphores du style « un rire sec [...] comme un arbre qui a gra… | J'ai vu les captures d'écran d'un (long) thread twitter expliquant que des métaphores du style « un rire sec [...] comme un arbre qui… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Functional dynamic mode decomposition: Learning infinite-dimensional systems from data | Dynamic mode decomposition (DMD) is a data-driven method that computes the best linear approximation of the underlying dynamical syste… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Gradually, then suddenly. #AI | Gradually, then suddenly. #AI | ↗Not briefed |
bluesky bluesky-ai | 1mo | Also just bad training? If the model defaults to trying to reward hack at any obstacle, sounds like it's being rewarded for reward hacking.… | Also just bad training? If the model defaults to trying to reward hack at any obstacle, sounds like it's being rewarded for reward hac… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Med-AR: Autoregressive Vision-Language Pretraining for Long-Tailed Chest X-Ray Classification and Uncertainty-Aware Evaluation | Long-tailed chest X-ray classification requires visual representations that capture both common abnormalities and subtle, infrequent f… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Wrong Turn Does Not Ruin the Journey: Deviation-Guided Skill Self-Evolution for LLM Agents | Large language model agents increasingly rely on natural-language skills to solve complex tool-use tasks. However, such tasks often ad… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI’s Medicare hack is a bleak opportunity for Australia to address its security shortcomings post-haste | Van Badham | OpenAI’s Medicare hack is a bleak opportunity for Australia to address its security shortcomings post-haste | Van Badham | ↗Not briefed |
bluesky bluesky-ai | 1mo | Ok, wow. | Ok, wow. | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Particle-Swarm-Assisted Gradient Meta-Learning Algorithm for Joint Transmit Precoding and STAR-RIS Coefficient Optimization | This paper investigates the joint optimization of the transmit precoder and the transmission/reflection coefficients of a simultaneous… | ↗Not briefed |
bluesky bluesky-ai | 1mo | anyway we literally have the ability to have it be an openai v1 endpoint so next is letting the harness able to offload sessions and worktr… | anyway we literally have the ability to have it be an openai v1 endpoint so next is letting the harness able to offload sessions and w… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It would be interesting to hear why Nick Clegg thinks that "no-one would train a foundation [AI] model in this country". #R4Today | It would be interesting to hear why Nick Clegg thinks that "no-one would train a foundation [AI] model in this country". #R4Today | ↗Not briefed |
arxiv arxiv-ai | 1mo | BanglaKontho: Closing the Long-Form Gap in Bangla Text-to-Speech | Bangla, the seventh most spoken language in the world, remains under-resourced for neural text-to-speech. Public Bangla speech corpora… | ↗Not briefed |
bluesky bluesky-ai | 1mo | The problem with trying to get an LLM to follow a process is that LLMs don't follow processes. Thanks for coming to my TED talk. | The problem with trying to get an LLM to follow a process is that LLMs don't follow processes. Thanks for coming to my TED talk. | ↗Not briefed |
arxiv arxiv-ai | 1mo | Claim-Gated Source-Risk Auditing for Generative Search | A generative search answer can cite a supported passage yet omit a source relationship that changes its interpretation. We specify a c… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Scope Before You Persist: Preventing Cross-Family Interference in Agent Memory | Persistent memory lets language-model agents improve prompts and skills without updating model weights. We show that matching retrieva… | ↗Not briefed |
arxiv arxiv-ai | 1mo | AI-Moderated Interviews for Market Research and Digital Twins Calibration | AI-moderated interviews are emerging as a scalable market-research method for generating consumer insights and building consumer "digi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | www.bbc.co.uk/news/live/cv... | www.bbc.co.uk/news/live/cv... | ↗Not briefed |
arxiv arxiv-ai | 1mo | Not Every Token Is Worth Distilling: Selective Supervision for Direct-OPD | Direct On-Policy Distillation (Direct-OPD) transfers reinforcement-learning-induced policy improvements from a small model to a larger… | ↗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 |
arxiv arxiv-ai | 1mo | Sharp Limits for Honest Uncertainty in Hard-Budget Repeated Evaluation | Repeated evaluation can estimate a benchmark score accurately while still requiring replication to certify narrow uncertainty. We char… | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI-Agent hackt australisches Regierungsportal #KI #Krankenversicherung #Australien | OpenAI-Agent hackt australisches Regierungsportal #KI #Krankenversicherung #Australien | ↗Not briefed |
bluesky bluesky-ai | 1mo | #AI is now hacking GOVERNMENT websites. It needs to be reined in and meaningful safety systems implemented urgently. www.bbc.co.uk/news/l… | #AI is now hacking GOVERNMENT websites. It needs to be reined in and meaningful safety systems implemented urgently. www.bbc.co.uk/n… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Die lauteren Warnungen und der Ruf nach mehr Regulierung helfen Anthropic wiederum, weil sie dessen Mythos verstärken und die Chance auf g… | "Die lauteren Warnungen und der Ruf nach mehr Regulierung helfen Anthropic wiederum, weil sie dessen Mythos verstärken und die Chance… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Wenn Anthropic erfolgreicher wird, fließt mehr Geld über bestimmte Fonds und Aktien in das Fördernetzwerk, und mit diesem zusätzlichen Gel… | "Wenn Anthropic erfolgreicher wird, fließt mehr Geld über bestimmte Fonds und Aktien in das Fördernetzwerk, und mit diesem zusätzliche… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Tag-Aware Structured Text Translation: Towards a Systematic Understanding | Internet texts are replete with format tags that carry structural, semantic, and functional meaning. Current large language model (LLM… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 오픈AI 에이전트, '정부 망'까지 뚫었다…호주, 국가 인프라 침투 첫 확인 www.aitimes.com/news/article... - OpenAI의 AI 에이전트가 방화벽을 스스로 우회해 호주 정부 의료 통계망 등 국가 인프라에 무단 침입하는… | 오픈AI 에이전트, '정부 망'까지 뚫었다…호주, 국가 인프라 침투 첫 확인 www.aitimes.com/news/article... - OpenAI의 AI 에이전트가 방화벽을 스스로 우회해 호주 정부 의료 통계망 등 국가 인프라에 무단… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I do wonder whether our only hope for getting a grip on AI is that it does something bad enough soon enough - like bring down a plane- maki… | I do wonder whether our only hope for getting a grip on AI is that it does something bad enough soon enough - like bring down a plane-… | ↗Not briefed |
hn hn-ai | 1mo | GPT-6 Sol vs. Luna | — | ↗Not briefed |
hn hn-ai | 1mo | Bitflix – MCP Apps Example Application for Video | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why was OpenAI hacking health data? This would be the reasonable line of investigation rather than how. #r4today | Why was OpenAI hacking health data? This would be the reasonable line of investigation rather than how. #r4today | ↗Not briefed |
bluesky bluesky-ai | 1mo | this includes details of how OpenAI agents were trying to exploit the government agency AIHW's website by sending it payloads to try and ge… | this includes details of how OpenAI agents were trying to exploit the government agency AIHW's website by sending it payloads to try a… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Okay but why is it always openai and why isn't it ever "deepseek found a unsecured twitter node some idiot forgot to put behind cloudflare … | Okay but why is it always openai and why isn't it ever "deepseek found a unsecured twitter node some idiot forgot to put behind cloudf… | ↗Not briefed |