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 | 4d | rotating "while LLM-related cognitive impacts are real […], there's also the potential [for craft]" around in my head | rotating "while LLM-related cognitive impacts are real […], there's also the potential [for craft]" around in my head | ↗Not briefed |
bluesky bluesky-ai | 4d | These styles are not accidental, and I know it sounds odd to critique a very impressive model on these grounds, but Anthropic is angling fo… | These styles are not accidental, and I know it sounds odd to critique a very impressive model on these grounds, but Anthropic is angli… | ↗Not briefed |
arxiv arxiv-ai | 4d | How Much of SAR Vehicle Recognition Is the Vehicle? Background Shortcuts, Angle Gaps and Partial Libraries on MSTAR and ATRNet-STAR | Deep networks score near 99% on the MSTAR vehicle benchmark, and the newer ATRNet-STAR benchmark was built to be harder. We ask how mu… | ↗Not briefed |
bluesky bluesky-ai | 4d | it seems to me like there is a broad ideological divide between people who tend to think the use of LLM agents is ultimately empowering vs.… | it seems to me like there is a broad ideological divide between people who tend to think the use of LLM agents is ultimately empowerin… | ↗Not briefed |
hn hn-ai | 4d | Evolving Infra: Agent-Adaptive LLM Serving at Scale | — | ↗Not briefed |
bluesky bluesky-ai | 4d | I fully believe that they only moderate via ai decisions lmao. When Claude is down, they pick a handful of trans folks to ban. | I fully believe that they only moderate via ai decisions lmao. When Claude is down, they pick a handful of trans folks to ban. | ↗Not briefed |
arxiv arxiv-ai | 4d | Urban Representation Learning for Identifying Potential Disparities in Crime Reporting | Open crime data support urban safety analysis but reflect reported incidents rather than the complete distribution of criminal activit… | ↗Not briefed |
rss the-decoder | 4d | Google warns that fully automated cyberattacks have arrived and will get worse fast | Google's Threat Intelligence Group has for the first time caught attackers finding and exploiting security vulnerabilities with no hum… | ↗Not briefed |
bluesky bluesky-ai | 4d | California Attorney General Rob Bonta files a sweeping lawsuit against OpenAI, alleging the technology giant jeopardized public safety by c… | California Attorney General Rob Bonta files a sweeping lawsuit against OpenAI, alleging the technology giant jeopardized public safety… | ↗Not briefed |
bluesky bluesky-ai | 4d | The Jensen Huang Soup Story is really fucking stupid and makes me hate him so much more, because no you worthless sack of pussycrazed shit,… | The Jensen Huang Soup Story is really fucking stupid and makes me hate him so much more, because no you worthless sack of pussycrazed… | ↗Not briefed |
arxiv arxiv-ai | 4d | TUSI: A Citation-Grounded Graph-Based Benchmark | Evaluating retrieval-augmented generation (RAG) systems in Persian law remains challenging because benchmarks lack realism, grounding,… | ↗Not briefed |
bluesky bluesky-ai | 4d | And this doesn't detract from the rather incredible results you can get from it. But I am now moving from "Anthropic is just that way" to "… | And this doesn't detract from the rather incredible results you can get from it. But I am now moving from "Anthropic is just that way"… | ↗Not briefed |
bluesky bluesky-ai | 4d | OpenAI steps up CSAM detection 🤖 OpenAI released new tools to combat child sexual abuse material, including an image classifier that flag… | OpenAI steps up CSAM detection 🤖 OpenAI released new tools to combat child sexual abuse material, including an image classifier that… | ↗Not briefed |
bluesky bluesky-ai | 4d | The classifier flagged fewer false positives for humans to review, while getting better at identifying abusive content. www.theverge.com/po… | The classifier flagged fewer false positives for humans to review, while getting better at identifying abusive content. www.theverge.c… | ↗Not briefed |
bluesky bluesky-ai | 4d | techcrunch.com/2024/07/24/m... | techcrunch.com/2024/07/24/m... | ↗Not briefed |
hn hn-ai | 4d | Show HN: Solipsist – Work with Pi agents using Markdown | Hi HN, since some time I have been experimenting with a new way to work with AI agents, instead of chatting with them in the usual sen… | ↗Not briefed |
rss gary-marcus | 4d | Is there a case for impeaching Trump over Jensengate? | We need to know more, but there’s definitely enough smoke now to warrant an investigation | ↗Not briefed |
bluesky bluesky-ai | 4d | I'd say they're saying the quiet part out loud, but it's not even quiet. They're just like 🤷🏻♂️ about betting the economy on Anthropic a… | I'd say they're saying the quiet part out loud, but it's not even quiet. They're just like 🤷🏻♂️ about betting the economy on Anthro… | ↗Not briefed |
bluesky bluesky-ai | 4d | "hey, that operating system you're using has LLM code in it 😏" 1. we know 2. not the point 3. also did you know that operating systems ar… | "hey, that operating system you're using has LLM code in it 😏" 1. we know 2. not the point 3. also did you know that operating syste… | ↗Not briefed |
bluesky bluesky-ai | 4d | i believe the way forward with anti-LLM action in software isn't a blanket boycott of anything that could possibly be affected, but instead… | i believe the way forward with anti-LLM action in software isn't a blanket boycott of anything that could possibly be affected, but in… | ↗Not briefed |
hn hn-ai | 4d | People will lose their jobs over GTA 6 | — | ↗Not briefed |
arxiv arxiv-ai | 4d | ARC-2: Continuous Compliance Through Delta Management | ARC 1 introduced a three-layer governance architecture for autonomous systems comprising model safety validation, cognitive certificat… | ↗Not briefed |
rss theverge-ai | 4d | Apple is testing daily usage limits for more complex Siri AI interactions | Apple released betas of iOS 27.3, iPadOS 27.3, and macOS 27.3 today, and the company is using the update to test new per device daily… | ↗Not briefed |
arxiv arxiv-ai | 4d | Evaluating the Evaluation of Music Generation Models with Generative Psychometrics | Subjective user studies are one of the major methods for evaluating music generation models. However, subjective evaluation protocols… | ↗Not briefed |
rss theverge-ai | 4d | OpenAI wants to stop deepfakes from overwhelming child abuse tiplines | OpenAI is announcing a tool that it says will help child sexual abuse investigators cut through the noise to find harmful material the… | ↗Not briefed |
arxiv arxiv-ai | 4d | SimAI: Mechanistic User Simulation through Psychological Network Models | Realistic user simulation enables the study of human interactions when direct experimentation is costly or potentially harmful. Howeve… | ↗Not briefed |
bluesky bluesky-ai | 4d | they're going to hate what happens when the Supreme Court rules that training an LLM is a transformative work | they're going to hate what happens when the Supreme Court rules that training an LLM is a transformative work | ↗Not briefed |
hn hn-ai | 4d | SignuDoc: Open-source document signature platform | — | ↗Not briefed |
bluesky bluesky-ai | 4d | I need these mfers to just tell me how to approve the form and let the publisher fight with its own imprint. I ended up copying what the ne… | I need these mfers to just tell me how to approve the form and let the publisher fight with its own imprint. I ended up copying what t… | ↗Not briefed |
arxiv arxiv-ai | 4d | Corrupting Ties: Investigating Robustness to Bias Injection Across Multi-Agent Network Topologies | AI systems are increasingly deployed in high-stakes decision-making contexts, where interactions among components can introduce fairne… | ↗Not briefed |
rss the-decoder | 4d | OpenAI's GPT-6 Astra tackles the Standard Model's toughest math problem on a lattice | OpenAI physicist Nathaniel Craig used GPT-6 Astra to tackle a problem that had stumped particle physicists for over 25 years. The AI m… | ↗Not briefed |
bluesky bluesky-ai | 4d | So should I skip the anthropic IPO?! | So should I skip the anthropic IPO?! | ↗Not briefed |
bluesky bluesky-ai | 4d | BREAKING: Researchers demonstrate 'sleeper agent' technique that embeds malicious payloads in open-weight AI models, activating only under … | BREAKING: Researchers demonstrate 'sleeper agent' technique that embeds malicious payloads in open-weight AI models, activating only u… | ↗Not briefed |
hn hn-ai | 4d | Show HN: Chisle – Claude Code plugin that cuts output and tool-output tokens | — | ↗Not briefed |
hn hn-ai | 4d | Brazil added to the list of countries with restricted US GPU imports | — | ↗Not briefed |
bluesky bluesky-ai | 4d | Everything is a cyber attack. Everything involves AI. The FAA grounding. The bank troubles. Me getting sick. It's all connected by Claude. | Everything is a cyber attack. Everything involves AI. The FAA grounding. The bank troubles. Me getting sick. It's all connected by Cla… | ↗Not briefed |
arxiv arxiv-ai | 4d | Data-Driven Variable-Length Compression of 3D Representations | Compressing 3D shapes is important for the efficiency of many applications, yet existing methods either cannot leverage dataset-level… | ↗Not briefed |
hn hn-ai | 4d | Show HN: CutCraft – an MIT library for making and editing videos in TSX | — | ↗Not briefed |
bluesky bluesky-ai | 4d | Have fun in San Francisco Superior, Sam | Have fun in San Francisco Superior, Sam | ↗Not briefed |
arxiv arxiv-ai | 4d | Chain-of-Thought Controllability Can Reduce Chain-of-Thought Monitorability | Chain-of-thought (CoT) monitoring is a promising tool for detecting hidden goals, off-task behavior, and other failures in AI agents.… | ↗Not briefed |
hn hn-ai | 4d | Nvidia, OpenAI and Anthropic to be named AI national champions, WSJ reports | — | ↗Not briefed |
hn hn-ai | 4d | How to build a collision-resistant hash function? – Merkle-Damgård | — | ↗Not briefed |
arxiv arxiv-ai | 4d | Can Coding Agents Write and Transpile Unit Tests | Coding assistants built on Large Language Models (LLMs) have become a staple of modern software development, increasingly trusted to e… | ↗Not briefed |
bluesky bluesky-ai | 4d | Meta and OpenAI want to take AI agents global, but fragile networks and language barriers pose steep hurdles across developing nations http… | Meta and OpenAI want to take AI agents global, but fragile networks and language barriers pose steep hurdles across developing nations… | ↗Not briefed |
hn hn-ai | 4d | California sues OpenAI over Hugging Face, other AI model loss of control🔧 Hugging Face | — | ↗Not briefed |
rss techmeme | 4d | Q&A with Satya Nadella on code verification becoming more important as code generation gets cheap, Microsoft's voluntary exit offers, decision models, and more (Gergely Orosz/The Pragmatic Engineer) | Gergely Orosz / The Pragmatic Engineer: Q&A with Satya Nadella on code verification becoming more important as code generation gets ch… | ↗Not briefed |
hn hn-ai | 4d | AI bubble will likely burst and cause a recession by 2028 election | — | ↗Not briefed |
arxiv arxiv-ai | 4d | Regularized Score Centering | Asynchronous reinforcement learning increases training throughput, but data from stale policies and numerical mismatch between trainin… | ↗Not briefed |
rss techcrunch-ai | 4d | Trump says ex-Nvidia CEO Jensen Huang’s affair with an alleged Chinese spy did not threaten national security | Trump contradicted his own government's intelligence officials, who reportedly concluded that Alice Huang is a Chinese spy. | ↗Not briefed |
bluesky bluesky-ai | 4d | Ruh roh Rorge 😀 www.latimes.com/california/s... | Ruh roh Rorge 😀 www.latimes.com/california/s... | ↗Not briefed |