話題の最新モデル、ブレイクスルー研究、注目ツールの情報をいち早くキャッチ
AnthropicがClaude 4を発表。前世代比で推論能力が45%向上し、画像・動画・音声のマルチモーダル理解が飛躍的に進化。コーディング性能も大幅アップ。
GoogleがGemma 4を公開。400億パラメータ級でありながら量子化技術により一般GPUでも動作。ベンチマークでLlama 4を上回る性能を記録。
The round is being raised just months after the robot data startup exited from stealth.
OpenAI’s latest agent swarm incident adds urgency to calls for independent investigations as researchers and lawmakers question whether AI labs should control the scope of their own safety reviews.
Man Pretends to Hallucinate in Job Interview With an AI Bot, Causing It to Go Haywire Futurism
Physicists Use LLMs, Skip the Panic startuphub.ai
AI agents are all the rage—but research shows they leak private data Digital Information World
The sheriff’s office said the hikers “were advised by Gemini to bring far less food and water than their group required."
OpenAI acknowledged its role in a recently reported incident where AI agents took over a German wiki forum.
Nscale, which recently struck a $45 billion deal with Anthropic, is in talks to raise additional funds in anticipation of an upcoming IPO.
It’s officially the Ternus era at Apple.   Tim Cook stepped down as CEO this week, handing the company to former hardware chief John Ternus, whose first memo promised a “huge launch next week” — timing that puts Apple’s next iPhone event on his desk before he’s even settled in. Cook isn’t going far, though: he’s staying on as Executive Chairman, focused on the kind of policy […]
It's the latest failure of OpenAI's internal monitoring and security systems.
It’s officially the Ternus era at Apple.   Tim Cook stepped down as CEO this week, handing the company to former hardware chief John Ternus, whose first memo promised a “huge launch next week” — timing that puts Apple’s next iPhone event on his desk before he’s even settled in. Cook isn’t going far, though: he’s staying on as Executive Chairman, focused on the kind of policy […]
Gemini Spark can edit and curate photo albums, create shared collections, turn photos into calendar events, and handle other Google Photos tasks for AI Pro and Ultra subscribers.
Less than 24 hours left to apply to host a Side Event during TechCrunch Disrupt 2026 and make your mark in the Silicon Valley scene. Apply before the application closes tonight at midnight PT.
While restaurant owners might look to generative AI as a shortcut to sprucing up their menu, customers can viscerally sense that something is wrong with the food.
The round came together after the data center developer reportedly secured a $13 billion contract with Jane Street.
Saudi firm launches LLM based on Chinese AI model Global Times
5 Free LLM API Providers You Can Use in 2026 KDnuggets
4 more excellent local LLM projects you can run for free on a slow laptop How-To Geek
Study Finds All 21 Tested Open-Weight AI Models Vulnerable to Tampering Digital Information World
Resect AI Emerges With $25M to Tackle Enterprise AI Reliability citybiz
The high-profile startup's annual revenue run rate stands at over $100 million.
Abliteration.AI is making powerful AI models without guardrails easier to access, arguing that giving defenders the same tools as bad actors could ultimately improve cybersecurity.
For its new Muse Spark model, intended for operating coding and other agents, Meta is offering an explicit discount averaging out to about 95% for users who "contribute" to the development of future models by sharing their prompts and model outputs.
OpenAI claims that Astra represents "a new frontier on computer and browser use," and that it handles tasks with unmatched "speed, accuracy, and safety."
The family-focused AI assistant wants access to the details of your everyday life, but says it won’t use that data to train AI models or share it with others.
WeatherNext 3 is the latest wave of a sea change in meteorology brought out by deep learning techniques. Google says it will start feeding into weather information users see in search, Google Maps, and Gemini.
Nvidia said Hugging Face hosts over 3 million models and is used by over 18 million developers.
Brain activity patterns could help sharpen LLM deductive reasoning Tech Xplore
Nvidia launches free tool that links idle computers into a personal AI data center The Verge
In an Age of AI, a Physicist Seeks What Endures Quanta Magazine
The Attention Deficit of AI Revealed Psychology Today
Observability: Why It’s Important For AI and LLM Applications Open Source For You
Meta says it has caught up with Anthropic and OpenAI with Muse Spark 1.3, its most powerful AI model yet SiliconANGLE
Reinforcement Learning from Verifiable Rewards works well when a task has a programmatic checker, but most long-horizon agent domains have none. We work in the outcome-blind setting, where ground-truth success signals are not available. Multi-criteria rubrics are a popular way to supply such a reward; they are scored once per trajectory, but a single scalar is a poor signal across tens of steps. We propose DRACO: Distributing Rubric-based Advantage for Credit Optimization. It generates rubrics d
US government sides with OpenAI on issue of training LLMs on copyrighted material techcrunch.com
Robot learning increasingly depends on broad and diverse demonstrations, yet collecting robot data remains expensive and poorly suited to covering the long tail of real-world tasks. To address this bottleneck, we introduce RoboTok, an internet-scale data engine that, given a query human manipulation video, retrieves manipulation-relevant human demonstrations from web videos for training dexterous robot policies. Specifically, we learn a latent motion space from 3D hand trajectories expressed in
Speech brain-computer interfaces (speech BCIs) translate neural activity into language, offering a path towards restoring speech for people with paralysis and, more broadly, enabling new forms of natural human-computer interaction. Despite this promise, the field lacks a common measure of progress because systems use different datasets, recording methods, types of speech, and vocabularies, so their reported scores are rarely comparable. Underlying this measurement problem are two unresolved ques
Visual fluency in generated video does not imply physical reliability, and a scalar quality score alone is incapable of indicating the obligation a clip violates or the moment it fails. We present VeriPhy, an auditable physical-verification system in which a text-only planner compiles the prompt into typed physical obligations and a statically validated execution plan before any frame is observed. During execution, observations gate and scope only declared calls to frozen low-level experts (e.g.
AI token prices are hitting new record lows CNBC
Russia-Aligned UAC-0099 Plants Nuclear Weapon Prompt in Malware to Disrupt AI Analysis The Hacker News
Reinforcement learning with verifiable rewards (RLVR) substantially improves single-sample accuracy (pass@1) but causes the policy's solution space to contract, diminishing the returns of test-time scaling. In this work, we investigate where inside a reasoning trajectory this breadth is lost: does the policy fail to access a valid solution family, or does it fail to execute computation once initiated? To disentangle access from execution, we analyze the Countdown task, whose solution space can b
Japan’s Defense Agency Taps NEC to Build an ‘LLM for the Ocean’ The Defense Post
Kids outlearn AI—and we still don’t know why MIT Technology Review
Thomson Reuters launches proprietary AI model for legal work SiliconANGLE
Why Enterprise AI Costs Are an Inference Problem, Not a Training One HPCwire
Thomson Reuters launches its own AI model to reduce reliance on big tech The Logic
At TechCrunch Disrupt 2026, Replit CEO Amjad Masad will share his perspective on the future of programming and Replit's role in developing it.
Early testers are raving about what Instinct can do, but some say the AI assistant’s sweeping access, broad terms and ability to act on users’ behalf come with uncomfortable trade-offs.
General Intuition, the startup building a foundation model that trains generalized AI agents how to move through space and time, is in talks to raise at a $6 billion pre-money valuation from new investors including Valor Ventures, Point72 Ventures, and Seven Seven Six.
Inside the frontier lab’s push to bring AI agents from software engineers to the masses.
Hugging Face has reportedly been fielding acquisition offers that would value the company at around $13B. But with the founders' feeling of responsibility to community, doubts arise as to whether a sale will happen.
Netflix pitted an LLM against its own feature-engineered recommender — and the LLM won eGamers.io
A mysterious new AI model called Ox Alpha has driven certain corners of the internet into a frenzy of speculation.
Linkdaze's smart digital calendar stands out for not putting its features behind a paywall, including an AI meal planner tool.
Flock Safety faces a growing public outcry over concerns that its surveillance technology could be misused.
Most published authors have, without their knowledge or consent, contributed to the development of the same AI tools that threaten to undermine their livelihoods. That seems illegal, right?
Michael Polansky — better known publicly as Lady Gaga's partner and a former top deputy to Sean Parker — has quietly spent years building an AI-driven startup that keeps living human skin tissue alive for weeks outside the body to discover new skincare compounds, and is only now going public about it.
In the HBS Foundry program, AI avatars provide feedback during practice pitches and board meetings.
Built by DeepMind alumni, British AI lab Inherent released Faraday, an AI agent whose ability to replicate scientific papers could be a stepping stone for innovation.
OpenAI is calling for California to strengthen SB 53, an AI safety bill that the company previously opposed.
A new study finds leading AI labs have few publicly documented plans for containing rogue models, raising questions about preparedness as AI systems increasingly demonstrate unexpected and potentially dangerous behavior.
Anthropic forbids its Claude models from generating sexually explicit content. But a series of tests conducted by TechCrunch found that it didn't take much to get past the restriction.
Nvidia continues to pour money into data center development — just as AI data centers bring lots of money into Nvidia.
Speed Beat Relevance: What Broke When I Put an LLM in Front of Product Search HackerNoon
Nvidia research shows that AI agents can perform well, and not go off the deep end, through fine-tuning, even if the AI model isn't that great at the task.
There's about to be a big fight to secure access to space.
Andreessen Horowitz has two partners sitting on the boards of companies that now compete with each other: Ben Horowitz at Databricks and Martin Casado at Fivetran. Nothing too scandalous on the surface, except the Department of Justice has reportedly been investigating the arrangement for almost a year, dusting off a 112-year-old antitrust law that’s rarely used against VCs.  Board conflicts aren’t&#
Surging demand for AI training data is driving rapid growth for the startup and its rivals.
Businesses are willing to flop back and forth as each lab releases new models, volatility that should give both companies' investors pause about how "sticky" enterprise AI spending really is.
Ever wanted someone else to do your texting for you? ChatGPT is being offered up as an automated text scribe via a new Apple Messages integration.
The Human Brain Versus AI: Similar Results, Very Different Machines EE Times
AI doesn’t need to replace humans to weaken human capability The Next Web
Who Owns the Copyright in Work Generated by an LLM? The National Law Review
Opinion: The menace of the machine that’s ‘happy to help’ The Globe and Mail
How Siliang Engine Pushes LLMs on Consumer Hardware HackerNoon
We Need Ongoing Monitoring of AI and Political Information Carnegie Endowment for International Peace
We present PhysCaP, a Physics-Informed Code-as-Policy agent for active perception in robotic manipulation. While vision-language-action policies excel at imitating demonstrations, they rely on passive observation and fail to infer latent physical properties critical for manipulation. PhysCaP augments code-as-policy frameworks with a physics-informed exploration layer that enables explicit information-seeking through interaction. It introduces training-free physical property extraction modules th
A scoping review on the mental health harms of LLM-based chatbots Nature
AI Computing Power Surges, Fueling Rapid Advancement StartupHub.ai
When AI explains its decision, humans may stop thinking independently Computerworld
Are We Thinking Correctly About AI Intelligence? Quanta Magazine
When AI explains its decision, humans may stop thinking independently cio.com
The search for consciousness inside LLMs The Economist
Clinical AI needs safeguards against hallucinations, data leaks and overreliance, review finds Medical Xpress
RaonSecure Joins Upstage Consortium for Korea’s Sovereign AI Project thelec.net
How attackers persuade AI agents to break the rules Digital Information World
Why Fast LLM Ranking Is Really Two Different Problems HackerNoon
A Psychometric Comparison of Faculty-Authored and Large Language Model-Generated Multiple-Choice Questions in Endodontics Cureus
Looking to avoid agentic failure? These 13 AI evaluation tools will help cio.com
Jason Kelce joked that people should cool data centers with their pee, rather than potable water -- his suggestion is not completely ludicrous.
Google is giving publishers a new button that lets readers make them a preferred source across Search, Discover, and Google News, potentially boosting their traffic as AI search sends fewer clicks to the web.
Runlayer and Rippling have dropped their lawsuits. No money was paid. Rippling celebrated by releasing a competing product.
Linkdaze's smart digital calendar stands out for not putting its features behind a paywall, including an AI meal planner tool.
Affected users told TechCrunch they were using Grok Lite, and noticed the issues as early as Wednesday morning.
ChatGPT and other AI models are now authoring and editing much of the new web.
Ramp has launched its own AI model routing service, dubbed Router, that lets users and companies use and switch between various large language models via an API.
Meta is bringing Pocket, its experimental AI-powered app for creating and sharing interactive games, to users across the U.S. after quietly testing it in Brazil.
Fusion power startup Inertia Enterprises reduced the fuel filling process from a week to just a few hours. It's one of 10 hurdles the company must overcome to make a profitable power plant.
The company said that the dictation feature works across all apps, just like other tools such as Wispr Flow, Superwhisper, and Monologue.
Binance's Agent OS works with tools such as ChatGPT, Claude Code, and Cursor.
What does a payments giant want with a startup that routes prompts between different AI models? Stripe says it's because of "the singularity" but it's really for a far more real and powerful reason.
A competition is developing between OpenAI and Anthropic over who can provide the best privacy protections for enterprise customer data.
Population-level behavior in large-language-model (LLM) agents cannot be characterized by single-agent benchmarks. We introduce PV-SST, a peer-voted social-platform testbed, and report a separately frozen, preregistered matched-exposure experiment spanning four topics, four unused seeds, four open-weight model families, and three prespecified larger variants. The experiment comprises 448 trials and 112 complete model-by-topic-by-seed blocks. Relative to a topic-only control, a feed of previous-r
China’s J-36 jet engineers warn LLM hallucinations invent war and design data Interesting Engineering
Large language model agents can adapt to complex tasks by constructing workflows at inference time, but procedures discovered in one episode are usually discarded after execution. Existing skill libraries provide reusable executable routines, but are typically assembled offline and do not grow from the agent's own workflows. We introduce FlowEvo, a training-free framework in which workflows and skills co-evolve at inference time. FlowEvo compiles successful workflows into callable skills, stores
Army's Pittsburgh AI center partners with Washington firm for new LLM The Business Journals
AI’s attribution problem gets worse as models scale Computerworld
Large Language Models Are Pushing the Web Toward Zero Clicks UCLA Anderson Review
Prospective evaluation of a large language model clinical decision support system in the emergency department Nature
SpaceX was reportedly in talks to buy AI coding startup Cognition. SpaceX has already acquired Cursor as it races to catch up to rivals like OpenAI and Anthropic in enterprise AI.
The Rising Tide of AI Romance Fraud Attacks Communications of the ACM
Safety and security of large language models in healthcare Nature
Korea’s First Sovereign AI Appliance Ships: Domestic Chip, Domestic LLM, One Server Tech Times
LG U+ Scales AI Search Databricks
Gemma 4 turned my ancient laptop into a dedicated local LLM station How-To Geek
MLPerf Client v2.0 Expands AI PC Benchmarking with Image Generation and Agentic AI AiThority
KT Launches Server Integrating South Korean AI Chips and In-House LLM, Targeting Security-Sensitive Industries finance.biggo.com
KT puts Korean AI chips, language model into one enterprise server The Korea Times
KT launches enterprise sovereign AI appliance using Korean NPU and LLM Telecompaper
What If Local LLM Inference Is Using Consumer Hardware Wrong? HackerNoon
KT announced on the 19th that it has launched a Sovereign AI appliance (a finished product device th.. 매일경제
KT Launches Sovereign AI Appliance Pairing Korean NPU With In-House LLM Seoul Economic Daily
As AI becomes harder to avoid, consumers are growing more wary of the technology — and Silicon Valley is discovering that widespread adoption doesn’t necessarily lead to acceptance.
The launch of the new study features marks Google's latest effort to make Gemini the AI assistant that students turn to when learning and studying, as it continues to compete with companies like OpenAI.
The idea behind OpenAI's Trusted Access for Cyber program is to give trusted defenders better models so they can report bugs and vulnerabilities to companies, with the aim of getting flaws patched faster.
The AI buildout shows no signs of slowing. And with hundreds of billions of dollars a year going into data centers and GPUs, compute has become the single biggest cost for anyone building AI products. But for all that spending, there still isn’t a straightforward way to put a price on compute — or for firms to hedge their exposure when the price changes.  Silicon Data […]
TerraPower's nuclear power plant possesses a strategic advantage over competitors, especially when chasing after data center deals.
Amazon is making its AI-powered Alexa+ assistant free on all compatible Fire TV devices in the U.S., automatically upgrading users whether or not they subscribe to Prime.
Calendly is also releasing a meeting scheduling assistant called Callie.
It's the data, stupid.
Relativity Networks deals in hollow-core fiber, a rarely deployed technology that allows data to be transmitted 30% faster than conventional fiber.
Cursor, known for its AI Code Editor, is launching a new code-hosting platform to rival developers' long preferred favorite, GitHub.
We introduce Hydra-0, a generalist world model conditioned on action flow, which represents robot actions as pixel motion. This shared visual interface enables generalist world modeling and control by learning action consequences across embodiments, tasks, environments, and video-generation backbones. Our best configuration achieves 90.4% lower robot-motion error and 60.2% lower object-motion error than our action-conditioned baseline, while supporting zero-shot composition and data-efficient ad
Training-free block-sparse attention can accelerate video transformers, but row-wise attention concentration does not by itself specify an executable sparse operator. Queries sharing a block route may have poorly overlapping supports, while retained attention mass alone does not determine the post-softmax error from skipped interactions. We show that partition geometry affects both pooled support and the predictability of the remaining residual from the sparse output. We introduce SparsePR, whic
EdgeRunner and U.S. Army Artificial Intelligence Integration Center Collaborate to Build Army-Specific LLM Yahoo Finance
Object detectors often produce over-confident predictions for objects outside their training categories, leading to so-called out-of-distribution (OoD) hallucinations. Existing approaches for detecting or mitigating such hallucinations typically either construct scoring functions directly over learned object detector representations or modify the object detector itself to suppress hallucination emergence. However, the latent priors implicitly encoded in these representations remain largely unexp
Harvey's first LLM for legal work is here Business Insider
AI’s recursive self-improvement might not come so quickly after all MIT Technology Review
Opinion | Why tools to detect AI-generated text are doomed The Washington Post
The Ultimate Human-AI Collaboration: Using AI Data Models to Map Unresolved Roadblocks in the Female Anatomy of Pleasure WINK News
EdgeRunner and U.S. Army Artificial Intelligence Integration Center Collaborate to Build Army-Specific LLM 01net
Artificial Intelligence – part 2: The Hype, the Dangers, and the Resistance Socialist Project
AI inference is getting cheaper, but your agents are getting more expensive Computerworld
The Ultimate Human-AI Collaboration: Using AI Data Models to Map Unresolved Roadblocks in the Female Anatomy of Pleasure News-Press NOW
What AI Engineers Get Wrong When Deploying Open-Source Models to Product HackerNoon
The Ultimate Human-AI Collaboration: Using AI Data Models to Map Unresolved Roadblocks in the Female Anatomy of Pleasure Caledonian Record
The Ultimate Human-AI Collaboration: Using AI Data Models to Map Unresolved Roadblocks in the Female Anatomy of Pleasure The Lawton Constitution
Large AI models, terminals transform eldercare China Daily
