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
17dSam Altman: You’re Completely Right About RAM. I’m Sorry for the Confusion. Here’s What Happened gizmodo.com/sam-altman-y...Sam Altman: You’re Completely Right About RAM. I’m Sorry for the Confusion. Here’s What Happened gizmodo.com/sam-altman-y...
↗Not briefed
bluesky
bluesky-ai
17danyways since they have "super intelligence" at their disposal why don't they just ask an LLM how to optimize their businesses in light of …anyways since they have "super intelligence" at their disposal why don't they just ask an LLM how to optimize their businesses in ligh…
↗Not briefed
bluesky
bluesky-ai
17dOpenAI just poached the product lead for Claude Code. I was wondering why a Claude Code product leader would leave pre-IPO Anthropic shar…OpenAI just poached the product lead for Claude Code. I was wondering why a Claude Code product leader would leave pre-IPO Anthropic…
↗Not briefed
bluesky
bluesky-ai
17dbut all they need to do to counter this gambit is to ask Claude how to adjust to the Dems being in power and sustain the AI boombut all they need to do to counter this gambit is to ask Claude how to adjust to the Dems being in power and sustain the AI boom
↗Not briefed
hn
hn-ai
17dBansat – Make your AI agent write the blueprint before the code—
↗Not briefed
bluesky
bluesky-ai
17dHis daughter now works for OpenAI, hired to lobby state governments on behalf of the AI oligarchs, and his former press secretary Josh Vlas…His daughter now works for OpenAI, hired to lobby state governments on behalf of the AI oligarchs, and his former press secretary Josh…
↗Not briefed
hn
hn-ai
17dAnthropic Needs to Disclose Its Tokenomics Before Its IPO—
↗Not briefed
rss
techcrunch-ai
17dApple is reportedly cutting budgets for its internal AI usageThe company is said to be significantly dropping the amount it's spending on AI for its engineering teams.
↗Not briefed
bluesky
bluesky-ai
17dt+7d: 24% fns/2.8% bc progress is noticeably slowing down - but that's okay, for I will do two things: - rework the processing pipeline t…t+7d: 24% fns/2.8% bc progress is noticeably slowing down - but that's okay, for I will do two things: - rework the processing pipel…
↗Not briefed
bluesky
bluesky-ai
17dOhio has an 8 gigawatt data center incoming from OpenAI. Voters went against all 21 measures there. Ballotpedia shows huge margins with on…Ohio has an 8 gigawatt data center incoming from OpenAI. Voters went against all 21 measures there. Ballotpedia shows huge margins wi…
↗Not briefed
bluesky
bluesky-ai
17dCat Wu, who left Anthropic last week, says she is joining OpenAI to lead product manangement for core products; Wu was the product lead for…Cat Wu, who left Anthropic last week, says she is joining OpenAI to lead product manangement for core products; Wu was the product lea…
↗Not briefed
bluesky
bluesky-ai
17dfair ok i'm gonna have to develop my own "make openai models happy" fork then :))))fair ok i'm gonna have to develop my own "make openai models happy" fork then :))))
↗Not briefed
rss
techmeme
17dCat Wu, who left Anthropic last week, says she is joining OpenAI to lead product manangement for core products; Wu was the product lead for Claude Code (Stephen Council/Business Insider)Stephen Council / Business Insider: Cat Wu, who left Anthropic last week, says she is joining OpenAI to lead product manangement for c…
↗Not briefed
arxiv
arxiv-ai
17dRobust Ranking of Port Impacts from Impending Tropical Cyclones: A Two-Stage Recommendation SystemWhen a tropical cyclone (TC) threatens several ports, preparedness planning depends on anticipating which ports are likely to be affec…
↗Not briefed
bluesky
bluesky-ai
17dit's hyperoptimized in ways specific to one frontier LLM family I'm really enjoying working with and I don't want it to be easily fingerpri…it's hyperoptimized in ways specific to one frontier LLM family I'm really enjoying working with and I don't want it to be easily fing…
↗Not briefed
arxiv
arxiv-ai
17dProvably Efficient Representation Learning for Low-Rank CMDPsWe study representation learning in low-rank Constrained Markov Decision Processes (CMDPs), where both value and reward functions are…
↗Not briefed
bluesky
bluesky-ai
17dSchumer needs to go for many reasons but in my mind one of the biggest is that his daughter is a lobbyist for OpenAI ( and for Amazon befo…Schumer needs to go for many reasons but in my mind one of the biggest is that his daughter is a lobbyist for OpenAI ( and for Amazon…
↗Not briefed
arxiv
arxiv-ai
17dHow do Transformers Perform Low-Level Control In-Context?One of the remarkable abilities of Transformer-based neural networks is in-context learning. This motivates us to think about "in-cont…
↗Not briefed
bluesky
bluesky-ai
17dHard to explain but the impact of Citrini flipping to bearish AI on Twitter is sort of like if Ed Zitron joined OpenAI as their head of com…Hard to explain but the impact of Citrini flipping to bearish AI on Twitter is sort of like if Ed Zitron joined OpenAI as their head o…
↗Not briefed
rss
techmeme
17dSources: Goldman Sachs and Apollo are in talks with investors to sell portions of SpaceX's ~$40B chip financing, asking some for up to $500M before syndication (Carmen Arroyo/Bloomberg)Carmen Arroyo / Bloomberg: Sources: Goldman Sachs and Apollo are in talks with investors to sell portions of SpaceX's ~$40B chip finan…
↗Not briefed
arxiv
arxiv-ai
17dHow Should Agents See Code? Rethinking Code Representations for Agentic Software EngineeringLLM agents are increasingly used for repository-level software engineering tasks. However, most agents still access repositories throu…
↗Not briefed
hn
hn-ai
17dAnthropic buys one-person Kendall Square biotech ahead of IPO. Here's why—
↗Not briefed
hn
hn-ai
17dShow HN: Exact integer LLL lattice basis reduction in C99LLL, also known as Lenstra–Lenstra–Lovász lattice basis reduction algorithm, is an important algorithm in mathematics and cryptography…
↗Not briefed
hn
hn-ai
17dCat Wu joins OpenAI from Anthropic in product-lead coup—
↗Not briefed
arxiv
arxiv-ai
17dEncoding and Propagation of Grammatical Number in Large Language ModelsSparse autoencoders (SAEs) have become an increasingly popular means of translating Large Language Model (LLM) hidden state vectors in…
↗Not briefed
arxiv
arxiv-ai
17dScenario Augmentation for Rare Financial Events via Multimodal Latent DiffusionExtreme market movements are scarce, heterogeneous, and difficult to learn from imbalanced historical data. We study whether generativ…
↗Not briefed
rss
techcrunch-ai
17dAfter the midterms, a bigger clash over data centers could be comingData center opposition crossed party lines in Tuesday’s midterms.
↗Not briefed
bluesky
bluesky-ai
17dSources: Anthropic and AP struck a licensing deal just before the midterms, marking the first time Anthropic paid a digital news publisher …Sources: Anthropic and AP struck a licensing deal just before the midterms, marking the first time Anthropic paid a digital news publi…
↗Not briefed
rss
techmeme
17dSources: Anthropic and AP struck a licensing deal just before the midterms, marking the first time Anthropic paid a digital news publisher to license content (Mark Stenberg/Adweek)Mark Stenberg / Adweek: Sources: Anthropic and AP struck a licensing deal just before the midterms, marking the first time Anthropic p…
↗Not briefed
arxiv
arxiv-ai
17dGraph Neural Networks for Resource Allocation in Integrated Sensing and CommunicationIn vehicular integrated sensing and communication (ISAC) networks, each service-provider vehicle uses a limited power budget to serve…
↗Not briefed
arxiv
arxiv-ai
17dEfficient Diffusion Policy Fine Tuning with Latent Noise Representation BridgingPre-trained diffusion and flow policies have emerged as powerful backbones for robotics, yet adapting them to novel data via reinforce…
↗Not briefed
bluesky
bluesky-ai
17dLOL. Cat Wu (a former Claude Code product manager) is joining OpenAI.LOL. Cat Wu (a former Claude Code product manager) is joining OpenAI.
↗Not briefed
arxiv
arxiv-ai
17dSatisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained targetAnimals and humans often do not maximize; they act relative to goals (aspiration levels) that are frequently not yet met. How can a go…
↗Not briefed
bluesky
bluesky-ai
17dI love how every time openai or whatever clamps down on some abuse of ai technology the boosters are like "people will just get around it d…I love how every time openai or whatever clamps down on some abuse of ai technology the boosters are like "people will just get around…
↗Not briefed
bluesky
bluesky-ai
17dif they were smarter the free press would train an LLM on his sermons & give him a column, so lucky usif they were smarter the free press would train an LLM on his sermons & give him a column, so lucky us
↗Not briefed
rss
techmeme
17dOpenAI launches Images 2.6 in ChatGPT, Codex, and the API, saying it generates images 50% faster, and adds animated GIF creation from prompts or photos (Emma Roth/The Verge)🔧 ChatGPTEmma Roth / The Verge: OpenAI launches Images 2.6 in ChatGPT, Codex, and the API, saying it generates images 50% faster, and adds anim…
↗Not briefed
hn
hn-ai
17dClaude / Codex is making a lot of us broke—
↗Not briefed
arxiv
arxiv-ai
17dHow Deep is the Signal? Time-Context up to 30 Years for Colorectal Cancer Prediction in a Primary-Care EHR ModelEHR foundation models read medical events within a fixed context window, and long context has so far been measured only in tokens. How…
↗Not briefed
hn
hn-ai
17dWhy LSP, Why the LSP Plugin, and Why Claude Code Doesn't Save Tokens—
↗Not briefed
arxiv
arxiv-ai
17dEnergy Measurement and Optimization of TinyML in Resource-Constrained Embedded Systems: A Systematic Mapping StudyThe rapid expansion of the Internet of Things (IoT) has increased the volume of data generated at the network edge, where many devices…
↗Not briefed
arxiv
arxiv-ai
17dHyFlash: A Prefix-Memory Prior for On-Device Speculative DecodingParallel block drafting accelerates speculative decoding by predicting multiple future positions in a single backbone pass, but on-dev…
↗Not briefed
bluesky
bluesky-ai
17d🙌 Community pick akii-seo-ai-search-optimizer — Free MIT-licensed Claude Code plugin for SEO, AEO and GEO. Audit websites, plan content, …🙌 Community pick akii-seo-ai-search-optimizer — Free MIT-licensed Claude Code plugin for SEO, AEO and GEO. Audit websites, plan cont…
↗Not briefed
hn
hn-ai
17dShow HN: Native HTML Partials and Native MFEsDear HN community,<p>I’ve been working on a specification and reference implementation called the Remote Module Definition (RMD) patte…
↗Not briefed
hn
hn-ai
17dAttackers used ChatGPT desktop and Claude Code in real intrusions🔧 ChatGPT—
↗Not briefed
bluesky
bluesky-ai
17dWho tf was saying it was a dream to work for OpenAI?Who tf was saying it was a dream to work for OpenAI?
↗Not briefed
bluesky
bluesky-ai
17dWe're going to have to liquidate OpenAI just to afford all of the plane tickets for the procession of artists carrying sama's head around (…We're going to have to liquidate OpenAI just to afford all of the plane tickets for the procession of artists carrying sama's head aro…
↗Not briefed
bluesky
bluesky-ai
17dThis is why I will bring everything local. Anthropic blocked Claude from using whisper to get the lyrics FROM ACTUAL MUSIC I OWN. Wow. Th…🔧 WhisperThis is why I will bring everything local. Anthropic blocked Claude from using whisper to get the lyrics FROM ACTUAL MUSIC I OWN. Wo…
↗Not briefed
rss
techmeme
17dSources: Google DeepMind spinout Isomorphic Labs is in talks to raise ~$5B at a ~$50B valuation; the AI drug discovery startup last raised $2.1B in May (Bloomberg)Bloomberg: Sources: Google DeepMind spinout Isomorphic Labs is in talks to raise ~$5B at a ~$50B valuation; the AI drug discovery star…
↗Not briefed
arxiv
arxiv-ai
17dReweighting is not Harmless: Understanding the Role of Dataset CharacteristicsReweighting is a widely used pre-processing method for promoting discrimination-free model behavior by assigning different weights to…
↗Not briefed
arxiv
arxiv-ai
17dTriAttention: Unified Sparse Mixed-Precision Attention for Long-Context LLM InferenceSparsity and quantization reduce the cost of attention by skipping interactions and lowering the precision of those retained. However,…
↗Not briefed