All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
The full TechCrunch Disrupt Stage lineup revealed, featuring Max Hodak, Mark Wahlberg, Benchmark partners, and more. Register now to join the conversations.
In 2026, the question for enterprise AI is no longer whether predictive models can outperform statistical forecasts—that argument is settled. The big question now is how to enable predictive systems to act on their own conclusions without drifting from business intent. The…
OpenAI's latest ad format will put images of sponsored products and services on your screen. The ads, which OpenAI will begin testing in the US later this month, will "initially" appear when you generate images with ChatGPT, according to an announcement on Monday. The company…
Через несколько минут мы узнаем, что OpenAI выпускает своего персонального ИИ-агента. Ну, а пока смотрите, что Meta(запрещена в РФ) сегодня днём представила персонального ИИ-агента для малого бизнеса. Это тот же Muse, но теперь он подключается к ресурсам…
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. People really hate AI, so why can’t they get enough? —Will Douglas Heaven Over the summer I talked to the CEO of Springboards, a startup…
Safeworld is building digital humans to make sure robots don't hurt the real ones.
🎨 GPT-6 Astra захватила сразу четыре лидерборда Design Arena — спустя всего месяц после релиза Модель OpenAI заняла первое место сразу в четырёх категориях: • 3D Design — 1484 • Frontend — 1397 • Full Stack — 1355 • Image-to-HTML — 1272 Особенно Astra отличилась в 3D-дизайне.…
OpenAI introduces a new visual ad format in ChatGPT and expands measurement tools, attribution partnerships, and brand suitability for advertisers.
Epoch AI: чипов, отгруженных к 2027г., может оказаться больше, чем спроса на ИИ-агентов. Но окупить их можно, если за агент-часы начнут платить как за работу человека. Epoch AI в своем свежем исследовании обсуждают вопрос: а сколько ИИ-агентов может работать одновременно на том…
Over the summer I talked to the CEO of Springboards, a startup building an LLM that’s designed to come up with a wider variety of responses than its mainstream rivals do. At the start of the call, he said something that’s been stuck in my head since: “We often say that we’re a…
Команда OpenAI встала на путь искупления. Теперь каждый день — либо действительно полезная функция, либо полный сброс лимитов. 🚬
🔥 Главное из мира ИИ за прошедшую неделю 💻Gemini 4 Argon — модель с генерацией ответов объёмом до миллиона токенов. 🌖GPT-6.1 Sol — обновление OpenAI с упором на код, документы и управление компьютером. 🔧Claude Sonnet 5.5 — новая модель Anthropic для кодинга и быстрых задач.…
Германия впервые выпустила модель. С результатом почти 97% в математике. Всего 78 млрд параметров, из которых на каждый токен активируются только 3,46 млрд. Открытые веса под лицензией Apache 2.0. Модель обучена для немецкого и английского языков. По собственным оценкам Aleph…
Streaming GPU solvers for entropic optimal transport (EOT), such as FlashSinkhorn, avoid storing the dense kernel but still evaluate all $n\times m$ point pairs in every Sinkhorn iteration. We present \textbf{FlashSinkhorn~2} (FS2), a solver for squared-Euclidean cost on…
Long-horizon research agents must decide both how to investigate and what to investigate next as evidence accumulates. This is hard to learn because such decisions are sparse in long execution traces, and their consequences may emerge several investigations later. We introduce…
In the UK, public healthcare staff report turning to machine translation (MT) - predominantly Google Translate (GT) - to communicate with patients across language barriers. Though intended to support their duty of care, potentially uninformed reliance on MT in such contexts…
Natural Visibility Graph (NVG) based analysis characterizes network traffic through topological descriptors reflecting different structural properties. However, not all descriptors contribute equally to cyber-attack classification, and extracting a large metric set can increase…
World models (WMs) simulate the transition dynamics of environments, enabling agents to plan over the consequences of their actions. In text-based environments, fine-tuning a Language Model (LM) to serve as a WM has emerged as a dominant paradigm. However, despite the widespread…
The operation of dynamically polarized targets in nuclear physics experiments relies on continuous tuning of the microwave frequency to compensate for radiation damage and evolving material properties, a task that is traditionally performed through manual trial-and-error by…
Developing cognitive models of algorithmic reasoning from behavioral data is a central problem in cognitive science that challenges current methods. Traditional approaches to cognitive modeling are interpretable and benefit from human expertise, but lack flexibility and…
Training-free safeguards for text-to-image generation often rely on a reusable safety signal, such as an unsafe direction or global toxic subspace, applied broadly across prompts. We provide a controlled geometric analysis of this global-unsafety assumption and reveal a…
Large language models spend the same amount of computation on every token they generate, regardless of how difficult each token is to produce. Methods such as speculative decoding and model routing are built on the premise that much of this computation is unnecessary, yet the…
World action models (WAMs) have emerged as a promising paradigm for robotic control by jointly predicting future visual dynamics and actions from an initial observation and instruction. However, existing WAMs struggle with long-horizon prediction, as generating dense video…
Societal resilience research relies on access to useful and actionable data, which motivates our main research question: Can annual reports, processed at scale with LLMs, provide a useful signal about how companies disclose their response to AI? We test this by applying a…
Large language models (LLMs) rely on long-horizon tool invocation sequences for complex tasks, where each invocation can alter the task state and condition subsequent decisions. In long-horizon tool use, final-outcome rewards provide weak credit assignment over long interaction…
Using a frontier model like Claude Opus as a meta-agent to generate terminal tasks and verifiers for RL training is increasingly common. Yet a runnable Docker image and executable test suite do not guarantee a faithful end-to-end pipeline for terminal agent training. We present…
Text-to-image generation enables users to explore several images generated from the same prompt. For these generated images to be useful, each one must reflect the user's preferences, measured by a learned reward, and differ visually from the others to maintain diversity.…
Recent years have seen the employment of a plethora of machine learning (ML) models in high-stakes domains, but they remain largely opaque to the practitioners who act on their predictions. While post-hoc explanation methods offer a lens into this model behavior, wielding them…
Interactive world models are increasingly capable of generating environments and acting within them, yet deliberately editing an existing executable world remains underexplored. We formulate world editing as intervening on an existing world while preserving properties that…
Campus AI tutors based on retrieval-augmented generation (RAG) must ground answers in assigned course materials while keeping textbooks and student dialogue on institutional infrastructure. We present CourseChat, an on-premises, multi-course RAG tutor for undergraduate business…