AI news archive.
19,864 articles, newest first, from the outlets listed on the sources page.
- arXiv cs.AIresearchThe Hallucination Signal Is a Mean Shift: Why Simple Probes Suffice
arXiv:2608.28930v2 Announce Type: replace-cross Abstract: Hidden-state probes effectively detect LLM hallucinations, but the geometry of the signal remains poorly characterized, driving increasingly complex probe archite…
- arXiv cs.AIresearchCPR for LLMs: Critical-Point Routing against Catastrophic Forgetting in Domain Adaptation
arXiv:2608.30158v4 Announce Type: replace-cross Abstract: Supervised fine-tuning (SFT) is the de facto standard for adapting large language models (LLMs) to target domains, but it often degrades the model's general capab…
- arXiv cs.AIagentsSIR: Self-improving Red-teaming for Compute Use Agents
arXiv:2608.30207v2 Announce Type: replace-cross Abstract: Computer-use agents (CUAs) are agents powered by vision-language models (VLMs) that perceive a screen and operate an operating system through mouse, keyboard, and…
- arXiv cs.AIresearchCapability-Gated Language Models: Security Composes, Utility Does Not
arXiv:2609.00445v3 Announce Type: replace-cross Abstract: Deployed language model safeguards (safety fine-tuning, filtering, unlearning) vary by principal only outside the model weights: filters are reconfigured, tiers a…
- arXiv cs.AIagentsFrom Fixed Keys to Readable Schemas: Small Language Models for Vehicle Agent Function Calls
arXiv:2609.09476v2 Announce Type: replace-cross Abstract: In-vehicle assistants must translate natural-language requests into accurate vehicle function calls under strict memory and latency constraints, making small lang…
- arXiv cs.AIresearchA Measured Communication-Quality Frontier for Federated LoRA Fine-Tuning with Adaptive Phase-Switching
arXiv:2609.13512v2 Announce Type: replace-cross Abstract: Federated fine-tuning of large language models with low-rank adaptation (LoRA) reduces the number of trainable parameters, but communication remains the dominant…
- arXiv cs.AIresearchTriCalRAG: A Three-Strategy, Retrieval-Augmented Benchmark for On-Premise LLM-Based Root Cause Analysis in AIOps
arXiv:2609.14762v2 Announce Type: replace-cross Abstract: Operational logs create a need for private, resource-efficient incident analysis, but aggregate detection scores can conceal severe prediction bias. We present Tr…
- SiliconANGLE AIpolicyUtah gives startup Nolla Health permission to start issuing AI automated prescriptions for acne treatments
Utah has become the first state in the U.S. to give an artificial intelligence startup permission to automate the approval of prescription refills with only minimal oversight from human physicians. In a blog post today,…
- LessWrongresearchTakes One to Know One - Training a model to grade reward hacks causes it to reward hack less itself
Code The question I asked in this project was - if I finetune a model to judge/catch reward hacks, does it change its behavior when completing the tests itself. Does it learn to hack more or less? Why it’s worth pursuing…
- LessWrongresearchYou can graft SDF changes from base models onto post-trained models
Webpage: praxis-research.org/grafting | Paper link: https://arxiv.org/abs/2610.00767 Work done during Peter and Dani's MATS Fellowship (10.0) under the mentorship of Shi, with Jinghua Ou as Research Manager. TLDR; We sha…
- TechRadar AIpolicyAmazon commits $1 billion to communities near data centers, AWS CEO says it no longer uses NDAs with government agencies
Mainstream AI news from a large tech publication.
- Guardian AIpolicyOpenAI has ‘work to do to rebuild trust’ in Australia, executive tells AI inquiry
Anthropic tells inquiry it ‘never tried to dictate’ Australian copyright rules as the ABC says copyright laws do not need changing Get our breaking news email , free app or daily news podcast OpenAI has used an appearanc…
- LessWrongpolicyThree types of AI risk: why (some) leftists are against (some) AI regulations
You may have seen factions of the left being opposed to current attempts to regulate the AI industry, despite the fact that they clearly also want to stop AI. I think I can explain why we're talking past each other. To d…
- LessWrongpolicyA closer look at what ozone and nuclear arms control tell us about regulating AI
Cross-posted from my Substack and adapted for LessWrong. Epistemic status: confident, with qualifications about what would change my mind at the bottom. An influential AI policy coalition, well-known to readers of this f…
- SiliconANGLE AIopen-sourceReflection AI debuts open-source Beam model with 501B parameters
Reflection AI Inc. today introduced Beam, an open-source large language model with 501 billion parameters. The launch comes a few months after the startup raised funding at a $25 billion valuation. Around the same time,…
- SiliconANGLE AIstartupsAI agent hardware startup Ghost, led by its 19-year-old-founder, raises $11M
Artificial intelligence hardware startup Ghost AI said today it has raised $11 million in funding to build what it believes is the first personal computer that’s dedicated to running personal AI agents. The round was led…
- TechRadar AIchipsGoogle and Microsoft hit by fresh criticism of their AI data center plans
Mainstream AI news from a large tech publication.
- Apple Machine Learning ResearchresearchRISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
Training a single LLM agent jointly across diverse interactive environments has attracted increasing attention as a route to generalist agents. Existing curriculum and data-selection strategies often allocate training at…
- Simon Willison's WeblogagentsQuoting Felix Rieseberg
The "old" version of Cowork runs model inference in the cloud, executing tool calls in an Anthropic-provided VM we shipped to your computer. We added the VM for capability, safety, and security reasons - mapping in just…
- LessWrongsafetyHumans Are Not Chinchillas: Revisiting “How Quick and Big Would A Software Intelligence Explosion Be?”
Introduction Tom Davidson and Tom Houlden’s prior model of an intelligence explosion poses the question: after AIs can entirely substitute for human researchers at the task of improving AI, how many year-equivalents of A…
- Engadget AImodelsGoogle removes free access to Gemini Flash and Pro on October 9
Free Gemini users are losing access to Google’s Flash and Pro AI models.
- LessWrongcommunityThe Unacknowledgable Labor of Pastor's Wives
I have another piece out in Asterisk, The Unacknowledgeable Work of Pastors’ Wives . This is a follow up to last year’s piece on church planting and the isomorphism to VC start-up culture. Discuss
- AWS Machine Learning BlogmodelsIntroducing GLM 5.3 on Amazon Bedrock
GLM 5.3 from Z.ai is now available on Amazon Bedrock: a 753B-parameter mixture-of-experts model built for coding and long-horizon agentic tasks. Learn how to invoke it with the OpenAI-compatible APIs, cut cost and latenc…
- The Verge AImodelsGemini Call for Me might tell your mom you’re running late
Google may be expanding its "Call for Me" AI feature beyond business calls so you can use it to send messages to friends and family. Android Authority reports finding a "Gemini Calling" introductory screen in an APK tear…
- LessWrongcommunityDance, Dance
As a brash but anxious young programmer in the before-time, I affected a standoffish attitude on the topic of games . Why would I learn to play a game as a human, I said, when I could just write a program to do it? When…
- SiliconANGLE AIchipsNetApp’s Novus tackles metadata bottlenecks in AI data infrastructure
AI data infrastructure must support growing GPU clusters while making enterprise data usable for models and agents. NetApp Inc. is tying intelligent data infrastructure to production AI through a unified approach to stor…
- Ars Technica AIsafetyMCP for agent-to-agent comms may be the riskiest protocol you’ve never heard of
Trust gaps in the new protocol spread malicious prompts from one agent to another.
- SiliconANGLE AIpolicyNetApp puts sovereign storage at the center of European infrastructure decisions
Sovereign storage is becoming an architectural and operational consideration for European enterprises as they invest in AI, modernization and cyber resilience. Sovereignty now covers much more than where data is physical…
- TechRadar AIsafetyOpen Microsoft database with 17 trillion total rows and 25,000 user accounts hacked by a bored teenager — but he's been well-paid for his actions
Mainstream AI news from a large tech publication.
- NVIDIA Developer BlogchipsControl How Your GPU Shares Work with Green Contexts
GPU applications increasingly consist of multiple independent components running at the same time within a single process: a latency-sensitive operator...
- TechRadar AIpolicyUS counties are desperately trying to stop data centers from taking over: A look at how Charlotte, Pima County, and Guildford County are dealing with the onslaught - with an interesting catch
Mainstream AI news from a large tech publication.
- TechCrunch AIpolicyOpenAI will start watermarking ChatGPT’s text in the EU
OpenAI will watermark ChatGPT and Codex text in the EU to comply with the AI Act. Editing can make the invisible marks harder to detect, it says.
- The Verge AIsafetyThis startup is issuing AI-generated acne prescriptions
People in Utah can now use AI to get a prescription for acne treatment. On Monday, healthcare startup Nolla Health announced that users in the state can scan their faces using its app, allowing its AI system to analyze a…
- LessWrongsafetyHow Would We Know? Reflections on trying to make AI go well amid deep uncertainty
Overview In this post, I try to make sense of live debates in AI safety. Then I explore the implications of my take on those debates for macrostrategy, and reflect on how the field thinks about and builds knowledge. My c…
- The Register AI + MLagentsCohere offers to put agents in lockdown mode with strict ACLs
Like OpenClaw but with access control
- TechCrunch AImodelsReflection debuts Beam, an open-weight AI model to rival Chinese models at lower compute cost
Reflection is aiming Beam and future models at enterprises and sovereign nations. The pitch is to build “AI factories,” a product that would let institutions build their own customized, local AI system by training Reflec…
- The Verge AIresearchAll the drama around AI’s takeover of mathematics
This past year, OpenAI, Anthropic, and other labs have announced breakthroughs on numerous long-standing mathematical problems, in some cases pushing well beyond what researchers expected current systems to be capable of…
- The Register AI + MLopen-sourceGet rid of Apple's AI bloatware and reclaim 12 GB of storage with this open source tool
Meet RemoveMacAI
- Ars Technica AIcommunityCommand-line tool quickly removes Apple Intelligence from macOS 27
The tool can free up over 12GB of storage.
- LessWrongsafetySelf-Modeling Interventions Modulate Emergent Misalignment
Code: https://github.com/atagade/sgtr-em Model checkpoints: https://github.com/atagade/sgtr-em/blob/main/MODEL_README.md Full paper (v2 coming soon): https://arxiv.org/abs/2606.23700 TL;DR A model’s self-model is its rep…
- The Verge AIagentsWikipedia operator says OpenAI’s ‘rogue’ bots may be linked to a May outage
Following many recent disclosures about AI agents accessing third-party websites and services, the Wikimedia Foundation, which hosts Wikipedia, says that it "can confirm that we have discovered some activity" by "rogue"…
- The DecodermodelsMeta and Microsoft pull back from Claude as Anthropic transforms from partner into competitor
Meta and Microsoft, two of Anthropic's biggest enterprise customers, are sharply cutting their use of Claude. Microsoft slashed the monthly per-employee budget in its cloud division from $100,000 to $10,000, while Meta h…
- TechCrunch AIagentsInstinct brings its AI agent to group chats, even for friends without an account
Instinct is launching group chats that let friends use its AI agent together for tasks like planning trips, organizing carpools, and coordinating events. The company says personal accounts remain separate, with permissio…
- NVIDIA Developer BlogchipsValidate AI Factory Changes with Digital Twins and AI Agents
AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration...
- Tom's Guide AIsafetyWhat is the ‘Hotel Lobby’ AI trend — and why is it raising concerns?
Consumer AI launches, tools, and practical coverage.
- LessWrongsafetyCurrent AIs out-persuade professionals in lab settings but (probably) not in the world
This article was created by Forethought . See all of our research on our website . How persuasive are present-day AI systems? Unfortunately, it is surprisingly difficult to know where AI capabilities in persuasion are to…
- TechRadar AIcommunitySurprise surprise, AI data centers are still incredibly unpopular across the US — and nearly two-thirds of Americans say they oppose new facilities
Mainstream AI news from a large tech publication.
- TechCrunch AIagentsTikTok rolls out an AI shopping assistant and one-click checkout
TikTok describes its new Shopping Assistant as a conversational AI agent designed to help users discover and purchase products.
- LessWrongsafetyResearch Note: Filtering Subversion-Relevant Information From Pretraining Data Is Feasible
Authors : Kyle O’Brien¹, Spencer Kitts², Cameron Tice¹, Alek Westover² ¹Geodesic Research, ²Redwood Research Figure 1: We pretrained LLMs from scratch with and without filtering subversion-relevant information , finding…
- The DecodermodelsReka AI's omni-model Rho-1 handles text, images, video, and robot control in a single model
Reka AI's Rho-1 is a 19-billion-parameter omni-model that processes and generates text, images, video, and robot control actions in a single neural network. Trained on 320 H100 GPUs in about three months, it uses a fract…