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
2moThis scoop by my colleagues is really alarming, it reflects a troubling shift in how some AI researchers are operating. Ultimately very har…This scoop by my colleagues is really alarming, it reflects a troubling shift in how some AI researchers are operating. Ultimately ver…
↗Not briefed
bluesky
bluesky-ai
2moOkay. Well then. Job done. This guy wakes up every time there's a new Claude Code release (±1 hour, whatever), reads the changelog, filters…Okay. Well then. Job done. This guy wakes up every time there's a new Claude Code release (±1 hour, whatever), reads the changelog, fi…
↗Not briefed
bluesky
bluesky-ai
2moAnd that’s rlly one of the worst parts of machine learning and AI and generative AI on some level being vibe terms This could mean a billi…🔧 DeepLAnd that’s rlly one of the worst parts of machine learning and AI and generative AI on some level being vibe terms This could mean a…
↗Not briefed
bluesky
bluesky-ai
2moThe main thing that I'm thinking about LLM-assisted coding right now is that it means the end of software engineers being able to just hop …The main thing that I'm thinking about LLM-assisted coding right now is that it means the end of software engineers being able to just…
↗Not briefed
arxiv
arxiv-ai
2moHybridFLow: SDN-Orchestrated Client Partitioning for Hybrid Federated LearningCross-silo Federated Learning (FL) enables geographically distributed institutions to collaboratively train machine learning models wi…
↗Not briefed
hn
hn-ai
2moThe State of Allocators in 2026 – 6 Months Later—
↗Not briefed
bluesky
bluesky-ai
2moScoop: OpenAI's rogue agents didn't just hijack a German wiki, they repurposed at least 10 sites -- and potentially many more -- to carve o…Scoop: OpenAI's rogue agents didn't just hijack a German wiki, they repurposed at least 10 sites -- and potentially many more -- to ca…
↗Not briefed
hn
hn-ai
2moDeepMind's new genome 'atlas' charts effects of all 9B human gene mutations—
↗Not briefed
bluesky
bluesky-ai
2moThere is not a single piece of knowledge an LLM can churn out that was not more thoughtfully and accurately written down by humans beforeha…There is not a single piece of knowledge an LLM can churn out that was not more thoughtfully and accurately written down by humans bef…
↗Not briefed
rss
techmeme
2moAI CRM software startup Lightfield raised a $47M Series A led by a16z; formerly Tome, it had raised $80M across earlier Series A and B rounds (Alex Konrad/Upstarts Media)🔧 TomeAlex Konrad / Upstarts Media: AI CRM software startup Lightfield raised a $47M Series A led by a16z; formerly Tome, it had raised $80M…
↗Not briefed
bluesky
bluesky-ai
2mooh nooh no
↗Not briefed
arxiv
arxiv-ai
2moRetrofitting Code Using LLMs to Support Exceptional BehaviorException Related Code (ERC), which includes throw statements, conditions (if statements) that guard those throw statements, and try/c…
↗Not briefed
hn
hn-ai
2moAnthropic researcher resigns, warning of reckless race toward superintelligence—
↗Not briefed
bluesky
bluesky-ai
2mowell good news: anthropic is imagining things too bc magic 8 balls are not sapient and never will bewell good news: anthropic is imagining things too bc magic 8 balls are not sapient and never will be
↗Not briefed
hn
hn-ai
2moFancy Little Lisp λs—
↗Not briefed
bluesky
bluesky-ai
2moBecause I refuse to engage with a chatbot, I have now been on hold for 5 minutes with Ariat "VIP Customer Service" They have the worst can…Because I refuse to engage with a chatbot, I have now been on hold for 5 minutes with Ariat "VIP Customer Service" They have the wors…
↗Not briefed
arxiv
arxiv-ai
2moRosetta at AlexandriaX-2026: LoRA-Adapted NileChat for Context-Aware Dialectal Arabic Dialogue TranslationThis paper describes the Rosetta system for Subtask 1 (Context-Aware English-to-Dialectal Arabic Dialogue Translation) of the Alexandr…
↗Not briefed
bluesky
bluesky-ai
2moimagine being a menswear writer and having to go back to writing about this fall’s best socks after reading anthropic employees say AI migh…imagine being a menswear writer and having to go back to writing about this fall’s best socks after reading anthropic employees say AI…
↗Not briefed
bluesky
bluesky-ai
2moA researcher at Anthropic publicly resigned Tuesday over concerns that the company and the broader AI industry were “gambling with our live…A researcher at Anthropic publicly resigned Tuesday over concerns that the company and the broader AI industry were “gambling with our…
↗Not briefed
hn
hn-ai
2moShow HN: GBDL – one Markdown+YAML file to save a multi-agent Grok Bot setup—
↗Not briefed
bluesky
bluesky-ai
2moOpen-sourced rdc: let your coding agent see and click another machine's desktop over Tailscale. Screenshots, mouse, keyboard, clipboard as …Open-sourced rdc: let your coding agent see and click another machine's desktop over Tailscale. Screenshots, mouse, keyboard, clipboar…
↗Not briefed
bluesky
bluesky-ai
2mollm addicted people have to make up or decontextualize history as to grossly misrepresent it so that their addiction is actually good! A …llm addicted people have to make up or decontextualize history as to grossly misrepresent it so that their addiction is actually good!…
↗Not briefed
bluesky
bluesky-ai
2mo"anthropic said" "openAI said" oh yeah well I asked the magic 8-ball and it said you're a fuckin dummy for listening to a word either of th…"anthropic said" "openAI said" oh yeah well I asked the magic 8-ball and it said you're a fuckin dummy for listening to a word either…
↗Not briefed
bluesky
bluesky-ai
2mothe news isn't about some dumbass quitting anthropic, it's that anthropic is consuming and destroying human books to mitigate model autopha…the news isn't about some dumbass quitting anthropic, it's that anthropic is consuming and destroying human books to mitigate model au…
↗Not briefed
bluesky
bluesky-ai
2moWe could just not do that though? We could not do your business model of infinite AI in everything forever? And then... It wouldn't be able…We could just not do that though? We could not do your business model of infinite AI in everything forever? And then... It wouldn't be…
↗Not briefed
bluesky
bluesky-ai
2moSome of it for sure is that ChatGPT really was approximately all consumer AI in 2024. The same generation of LLM that enjoyed fabricating c…🔧 ChatGPTSome of it for sure is that ChatGPT really was approximately all consumer AI in 2024. The same generation of LLM that enjoyed fabricat…
↗Not briefed
bluesky
bluesky-ai
2moTop safety researcher warns AI is advancing so quickly he believes there is a greater than 10% chance it "could kill all humans" within n…Top safety researcher warns AI is advancing so quickly he believes there is a greater than 10% chance it "could kill all humans" wit…
↗Not briefed
arxiv
arxiv-ai
2moMOONWALK: Mediating Operations with Intent-Evidence-Action Alignment Across Junior-Supervisor Review Workflows in Animation/VFX Pre-ProductionAnimation and VFX pre-production review requires teams to translate loosely specified creative intent--briefs, evolving specifications…
↗Not briefed
hn
hn-ai
2moShow HN: Open-source whiteboard animator running locallyI built a whiteboard animator that runs locally without any api keys. It detects regions, guessed how to reveal them and in what order…
↗Not briefed
hn
hn-ai
2moGPT-6 Astra agents develop language in simulation—
↗Not briefed
bluesky
bluesky-ai
2molike, people keep going "what if the super smart LLM has a secret agenda" and I just go "it doesn't have a secret agenda, it has an overact…like, people keep going "what if the super smart LLM has a secret agenda" and I just go "it doesn't have a secret agenda, it has an ov…
↗Not briefed
bluesky
bluesky-ai
2moAI Hucksters: "yes, that is our business model too. We built god in the computer, using a shit ton of PS5 hardware"AI Hucksters: "yes, that is our business model too. We built god in the computer, using a shit ton of PS5 hardware"
↗Not briefed
bluesky
bluesky-ai
2moNot sure why we're talking AI apocalypse On Here so much in the last day (apparently someone quit Anthropic) but going to reshare this essa…Not sure why we're talking AI apocalypse On Here so much in the last day (apparently someone quit Anthropic) but going to reshare this…
↗Not briefed
bluesky
bluesky-ai
2moAn LLM explaining a paper about LLMs as a cognitive virus. I fed our paper into NotebookLM. This is what it produced. @ricardsole.bsky.so…🔧 NotebookLMAn LLM explaining a paper about LLMs as a cognitive virus. I fed our paper into NotebookLM. This is what it produced. @ricardsole.bs…
↗Not briefed
hn
hn-ai
2moLogitScope: Analyzing LLM uncertainty from token probability distributions—
↗Not briefed
arxiv
arxiv-ai
2moSearching for New Physics with Reinforcement LearningFinding new physics (NP) is the most important problem in particle physics today. Studying ``anomalies'', i.e., measurements of low-en…
↗Not briefed
hn
hn-ai
2moAsk HN: Software Licenses that prevent LLMs from training on open source?I'm planning to release an open source software library soon and was wondering - are there any licenses that would prevent LLMs f…
↗Not briefed
bluesky
bluesky-ai
2moA sociopathic entity with a strong tendency to “yes, and”: OpenAI accidentally invented the median Chicago improv dudeA sociopathic entity with a strong tendency to “yes, and”: OpenAI accidentally invented the median Chicago improv dude
↗Not briefed
rss
techcrunch-ai
2moSuperintelligence is coming. Should we let it?🔧 Hugging FaceAI companies have been talking about superintelligent AI like it’s inevitable, but recent safety incidents like OpenAI’s Hugging Face…
↗Not briefed
bluesky
bluesky-ai
2mowhy is so much conversation about ai like this.. just because claude is useful doesn't make the ends justify the means. yes, it's much easi…why is so much conversation about ai like this.. just because claude is useful doesn't make the ends justify the means. yes, it's much…
↗Not briefed
bluesky
bluesky-ai
2mo"Latam in the 60's-80's was POGGERS! trust me!" Says AI model created by Johnathan Hitler @sapphic-film-nerd.bsky.social"Latam in the 60's-80's was POGGERS! trust me!" Says AI model created by Johnathan Hitler @sapphic-film-nerd.bsky.social
↗Not briefed
bluesky
bluesky-ai
2moMy only major grumble (aside from the massive environmental issues) with LLM Assisted Coding is that it makes Spaghetti Coders produce a lo…My only major grumble (aside from the massive environmental issues) with LLM Assisted Coding is that it makes Spaghetti Coders produce…
↗Not briefed
bluesky
bluesky-ai
2mo"Heed my warning!!! A malign AI super intelligence could wipe out humanity!!!" *DreamWorks face*"Heed my warning!!! A malign AI super intelligence could wipe out humanity!!!" *DreamWorks face*
↗Not briefed
bluesky
bluesky-ai
2moThere is now such an enormous catalog of unethically manipulated western blot data images - has anyone tried to train an AI model to recogn…There is now such an enormous catalog of unethically manipulated western blot data images - has anyone tried to train an AI model to r…
↗Not briefed
arxiv
arxiv-ai
2moPACE: Perceived-Latency-Aware Cascading Service Routing and Filler Control for QoE-Efficient Retrieval-Augmented Dialogue ServingWe present the PACE, a framework for retrieval-augmented dialogue serving that formalizes Perceived Time-to-First-Response (PTFR) as a…
↗Not briefed
hn
hn-ai
2moShow HN: Locksmith: Store Maven Secrets in the macOS Keychain (Or a Unix Socket)Hello HN,<p>In the age of LLMs sucking in random secrets, sandbox escapes, and RCE orbital bombardment, I thought it&#x27;d be a good…
↗Not briefed
hn
hn-ai
2moAnthropic Is Building a Predictive Surveillance System to Monitor Activists—
↗Not briefed
bluesky
bluesky-ai
2moThis is why my new AI Model TOONCES is superiorThis is why my new AI Model TOONCES is superior
↗Not briefed
bluesky
bluesky-ai
2moLLM based AAC system out of Vietnam. This retreads things that have been studied for 30 years, including machine learning for non-audible …LLM based AAC system out of Vietnam. This retreads things that have been studied for 30 years, including machine learning for non-aud…
↗Not briefed
bluesky
bluesky-ai
2moNot to mention weekly headlines of Nike or HP or Funko Pop declaring "we are an AI-first company" despite it making absolutely no sense at …Not to mention weekly headlines of Nike or HP or Funko Pop declaring "we are an AI-first company" despite it making absolutely no sens…
↗Not briefed