AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
bluesky
bluesky-ai
2moOpenAI chief executive Sam Altman said that the world is “right to be afraid” about rapidly advancing artificial intelligence, but that ind…OpenAI chief executive Sam Altman said that the world is “right to be afraid” about rapidly advancing artificial intelligence, but tha…
↗Not briefed
hn
hn-ai
2moWASM to Go Source Code—
↗Not briefed
hn
hn-ai
2moAsk HN: How would you anthropomorphize the Anthropic Claude logo?As the title states.
↗Not briefed
bluesky
bluesky-ai
2moConcerns about how quickly and recklessly AI companies are developing their tech *should* point us in the direction of regulating the indus…Concerns about how quickly and recklessly AI companies are developing their tech *should* point us in the direction of regulating the…
↗Not briefed
rss
theverge-ai
2moThe AI data center e-waste problem is huge — and getting biggerE-waste from the AI boom has been vastly underestimated, a new report warns. By 2050, it could become enough trash to fill 23 million…
↗Not briefed
bluesky
bluesky-ai
2mo@sentinelone.com also has a report out by @hegel.bsky.social on the same trail of evidence spotted by Jonas Wiedermann-Möller: www.sentinel…@sentinelone.com also has a report out by @hegel.bsky.social on the same trail of evidence spotted by Jonas Wiedermann-Möller: www.sen…
↗Not briefed
bluesky
bluesky-ai
2moAgreed. They inaccurately describe their "reasoning", so there is no way to determine it. They are statistical sieves, word predictors. …Agreed. They inaccurately describe their "reasoning", so there is no way to determine it. They are statistical sieves, word predicto…
↗Not briefed
bluesky
bluesky-ai
2moEs más, me juego la vida contra un vermú a que la casualidad de los ataques por "IAs descontroladas" de las últimas semanas, justo a 1 puto…Es más, me juego la vida contra un vermú a que la casualidad de los ataques por "IAs descontroladas" de las últimas semanas, justo a 1…
↗Not briefed
hn
hn-ai
2moDOE vs. GitHub, INC: LLM generated-content not a DMCA violation [pdf]—
↗Not briefed
hn
hn-ai
2moClaude.md is good for taste and project context.It's a weak place for invariants—
↗Not briefed
bluesky
bluesky-ai
2moExperts have spent decades hunting for a special class of ever more complex elliptic curves. The last breakthrough took mathematicians 18 y…Experts have spent decades hunting for a special class of ever more complex elliptic curves. The last breakthrough took mathematicians…
↗Not briefed
arxiv
arxiv-ai
2moEfficient Nash Equilibrium Computation for Cybersecurity GamesComputing Nash equilibria of simulation-based cybersecurity games with policy-space response oracles (PSRO) is bottlenecked by payoff…
↗Not briefed
arxiv
arxiv-ai
2moA Cross-Lingual Acoustic Disease-Alignment Framework for Respiratory Health Assessment from Spontaneous SpeechSpontaneous speech offers a scalable, noninvasive signal for respiratory health assessment, yet interpretable models that generalize a…
↗Not briefed
bluesky
bluesky-ai
2moYesterday on NPR, Anthropic co-founder Jack Clark: “show awareness” “persuaded one another” “hacked their way out” communicated… worked to…Yesterday on NPR, Anthropic co-founder Jack Clark: “show awareness” “persuaded one another” “hacked their way out” communicated… work…
↗Not briefed
hn
hn-ai
2moMozilla and Mistral partner to expand AI competition and preserve user choice—
↗Not briefed
bluesky
bluesky-ai
2moOpenAI, Anthropic, and the major Chinese labs all do this already. It's how their cyber classifiers work. But you don't have to use their i…OpenAI, Anthropic, and the major Chinese labs all do this already. It's how their cyber classifiers work. But you don't have to use th…
↗Not briefed
bluesky
bluesky-ai
2moThe thing is, I understand that with a good chunk of FOSS right now you're going to run into the odd LLM contribution. I don't like it, but…The thing is, I understand that with a good chunk of FOSS right now you're going to run into the odd LLM contribution. I don't like it…
↗Not briefed
hn
hn-ai
2moShow HN: Plexus – Telemetry for hardware teams who could build it but shouldn'tHi, I&#x27;m Ann, building Plexus.<p>We gives hardware teams live dashboards and stored history for your sensor data, so you don&#x27;…
↗Not briefed
bluesky
bluesky-ai
2moI do think the AI doomer stuff is in a superposition of goofy and not goofy. the universal paperclips stuff is pretty goofy. the "some dumb…I do think the AI doomer stuff is in a superposition of goofy and not goofy. the universal paperclips stuff is pretty goofy. the "some…
↗Not briefed
bluesky
bluesky-ai
2moAnthropic yksinkertaistaa tekoälytyökalunsa Clauden toimintaa ja yhdistää Claude Chatin ja Coworkin Jatkossa Claude osaa laajentaa vaikkap…Anthropic yksinkertaistaa tekoälytyökalunsa Clauden toimintaa ja yhdistää Claude Chatin ja Coworkin Jatkossa Claude osaa laajentaa va…
↗Not briefed
bluesky
bluesky-ai
2moActually while I've got you on the phone, if the robot is so smart, can't we use the robot? Maybe all queries ought to be put through via a…Actually while I've got you on the phone, if the robot is so smart, can't we use the robot? Maybe all queries ought to be put through…
↗Not briefed
bluesky
bluesky-ai
2moAI spots fake URLs with almost perfect accuracy 🔗 A new hybrid deep-learning model identifies spoofed website addresses with up to 99% ac…AI spots fake URLs with almost perfect accuracy 🔗 A new hybrid deep-learning model identifies spoofed website addresses with up to 9…
↗Not briefed
bluesky
bluesky-ai
2mo‘Two academics threw cold water on the idea that the Hugging Face swarm hack by OpenAI agents occurred because of a breakthrough in intelli…🔧 Hugging Face‘Two academics threw cold water on the idea that the Hugging Face swarm hack by OpenAI agents occurred because of a breakthrough in in…
↗Not briefed
bluesky
bluesky-ai
2moAI is going to kill us all because no one bothered to take the remote control. Too busy doom scrolling about the AI doom rhetoric. Anthro…AI is going to kill us all because no one bothered to take the remote control. Too busy doom scrolling about the AI doom rhetoric. A…
↗Not briefed
bluesky
bluesky-ai
2moMitch McConnell dispels rumors about his death just in time for Anthropic insiders to announce that AI could kill us all in as little as si…Mitch McConnell dispels rumors about his death just in time for Anthropic insiders to announce that AI could kill us all in as little…
↗Not briefed
bluesky
bluesky-ai
2moBreaking : les « woke wizards of AI » rémunérés par Anthropic sont aussi capitalistes que l’administration Trump. Ils n’ont aucun désaccord…Breaking : les « woke wizards of AI » rémunérés par Anthropic sont aussi capitalistes que l’administration Trump. Ils n’ont aucun désa…
↗Not briefed
bluesky
bluesky-ai
2moThis is the AI dev environment I built in December... This video from January shows adding some new fields and a new module to a formal #ga…This is the AI dev environment I built in December... This video from January shows adding some new fields and a new module to a forma…
↗Not briefed
arxiv
arxiv-ai
2moMAGS: Multi-agent Auto-formalization Guarantees Safety for Agentic OutputsLLM coding agents now generate complex programs at a scale that makes thorough human review increasingly difficult, raising the risk o…
↗Not briefed
bluesky
bluesky-ai
2moOh it was gonna talk deepseek with you but you meant…okOh it was gonna talk deepseek with you but you meant…ok
↗Not briefed
bluesky
bluesky-ai
2moOpenAI says CEO Altman will attend Trump's state dinner for Xi reut.rs/4AoppJhOpenAI says CEO Altman will attend Trump's state dinner for Xi reut.rs/4AoppJh
↗Not briefed
hn
hn-ai
2moDriftsort: An efficient, generic and robust stable sort implementation—
↗Not briefed
bluesky
bluesky-ai
2moHighlighting this from the Anthropic risk report: "We identified a PRC government-aligned operation that used Claude to track, profile, and…Highlighting this from the Anthropic risk report: "We identified a PRC government-aligned operation that used Claude to track, profile…
↗Not briefed
rss
techcrunch-ai
2moAfter accusations of selling ‘perv glasses,’ Meta prepares to sell a pair without a cameraCan Meta dodge the "pervert glasses" accusations with a new camera-free product?
↗Not briefed
arxiv
arxiv-ai
2moDo AI Agents Understand Computer Architecture?Agents are increasingly asked to design hardware, and increasingly reported to succeed. Such reports establish that a design improved;…
↗Not briefed
bluesky
bluesky-ai
2mo"Us anunciem que la nostra empresa té molts beneficis si tenim en compte aquesta nova manera de calcular les finances basada en ignorar tot…"Us anunciem que la nostra empresa té molts beneficis si tenim en compte aquesta nova manera de calcular les finances basada en ignora…
↗Not briefed
bluesky
bluesky-ai
2mounironically if you're using any kind of ai filter thing like dlss5 on your VTUBER MODEL I hope your artists blacklist you within their cir…unironically if you're using any kind of ai filter thing like dlss5 on your VTUBER MODEL I hope your artists blacklist you within thei…
↗Not briefed
hn
hn-ai
2moAsk HN: Where Are All the Sysadmins?Been genuinely trying to hire someone competent that can deliver&#x2F;manage HA solutions on-prem based on open source software, for m…
↗Not briefed
hn
hn-ai
2moShow HN: Halo 3's Guardian, playable in the browser, as my personal website🔧 ReplicateI&#x27;ve never had a personal website, and I didn&#x27;t want to make a resume&#x2F;portfolio site, so I rebuilt Guardian and hid a t…
↗Not briefed
bluesky
bluesky-ai
2moonce again setting a pre-2022 filter on my search because if you just look for hojicha latte recipes they're all written in horrible LLM ca…once again setting a pre-2022 filter on my search because if you just look for hojicha latte recipes they're all written in horrible L…
↗Not briefed
bluesky
bluesky-ai
2moIs the AI industry just trolling itself at this point?Is the AI industry just trolling itself at this point?
↗Not briefed
hn
hn-ai
2moMCP server that gives any agent a sandbox, browser and GitHub access—
↗Not briefed
bluesky
bluesky-ai
2moIt's crazy to me that people who falsely champion the garbage LLM crap as some sort of gateway that circumvents elitism and suppression of …It's crazy to me that people who falsely champion the garbage LLM crap as some sort of gateway that circumvents elitism and suppressio…
↗Not briefed
arxiv
arxiv-ai
2moMammography Foundation Models for Opportunistic Prediction of Major Adverse Cardiovascular EventsCardiovascular disease (CVD) remains the leading cause of death among women, yet cardiovascular risk assessment often relies on clinic…
↗Not briefed
arxiv
arxiv-ai
2moRiemannian--Lorentz Fusion of Vision Transformers and State-Space ModelsScaling deep learning faces critical bottlenecks: data exhaustion, exponential training costs, and resource concentration. Model mergi…
↗Not briefed
rss
techmeme
2moHang Ten Systems, which uses AI to help large enterprises build software, raised an additional $53M seed led by Xora, five weeks after its initial $32M seed (Jagmeet Singh/TechCrunch)Jagmeet Singh / TechCrunch: Hang Ten Systems, which uses AI to help large enterprises build software, raised an additional $53M seed l…
↗Not briefed
bluesky
bluesky-ai
2moSorry to add fo this but DLSS5 is, by their own admission, gen-AI. This comes directly from NVIDIA's website. I would hate to see any vtube…Sorry to add fo this but DLSS5 is, by their own admission, gen-AI. This comes directly from NVIDIA's website. I would hate to see any…
↗Not briefed
bluesky
bluesky-ai
2moThat depends on the government. If these guys get their legal moratorium on AI model training, everyone crashed around them, but they're f…That depends on the government. If these guys get their legal moratorium on AI model training, everyone crashed around them, but they…
↗Not briefed
arxiv
arxiv-ai
2moFCx: An algorithm for finding Feasible Counterfactual ExplanationsCounterfactual (CF) explanations identify changes that alter an input's classification. While existing methods produce realistic and l…
↗Not briefed
bluesky
bluesky-ai
2mo"At a minimum, the U.S. government should work with frontier labs to ensure that they are deploying robust real-time monitoring processes t…"At a minimum, the U.S. government should work with frontier labs to ensure that they are deploying robust real-time monitoring proces…
↗Not briefed
bluesky
bluesky-ai
2mook, bidding for my half-finished novel will open at a bargain-basement $20 billion www.nytimes.com/2026/09/16/b...ok, bidding for my half-finished novel will open at a bargain-basement $20 billion www.nytimes.com/2026/09/16/b...
↗Not briefed