All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
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. Open models have real benefits. Do they outweigh the risks? Anthropic recently released a report arguing that a powerful new model released…
Anthropic: very soon small AI models will be able to self-replicate on smart ovens, TV boxes, etc. without access to data centers The authors studied autonomous replication and adaptation (ARA) - this is when the model itself deploys and distributes without human intervention,…
Anthropic says it has cut the extraction of Claude's internal reasoning tokens by 90 to 99 percent. The company links the large-scale distillation attacks to Chinese companies Zhipu and Moonshot, and has already banned tens of thousands of accounts. The article Anthropic says it…
🚫 Anthropic made copying Claude 100 times more difficult and blocked tens of thousands of accounts. Anthropic stated that it has implemented new mechanisms to protect Claude from distillation — a method where responses from a powerful AI model are used to train another neural network. According to the company's estimate,…
Small AI models may be just months away from spreading on their own across poorly secured consumer devices like smart TVs and robot vacuums, according to a new Anthropic study. In tests, one model already completed the full infection chain. CEO Dario Amodei has warned that a…
Anthropic researchers tested whether small AI models could replicate themselves and spread across a network of devices.
From public services to robotics: how Kazakhstan develops the digital economy Profit.kz
MIT Technology Review Explains: Let our writers untangle the complex, messy world of science and technology to help you understand what’s coming next. You can read more from the series here. Late last month, Anthropic released a report arguing that the model GLM-5.3 poses a…
Something is changing in the world of mathematics and science and it is changing fast. Open AAI just released hundreds of AI generated mathematical results including claims that would have sounded completely impossible just a few months ago. I've been scrolling so much here.…
✔️ Anthropic blocked more than 11 million accounts in six months In an updated transparency report, Anthropic reported that it blocked 11.4 million accounts in the first half of 2026. This is almost 8 times more than in the second half of 2025, when there were 1.45 million blockings...
🧐 Zuckerberg refused to slow down the development of AI: Meta intends to accelerate the race with OpenAI and Anthropic Mark Zuckerberg said that Meta has finally reached an advanced level in the development of artificial intelligence and is now going to move even faster. The statement came against the backdrop...
France is actively developing agent computing using NVIDIA technologies.
Leonid Vinyar, one of the founders of the Karaganda IT company Plusmicro, has died. EPAM Systems acquired it in 2008. We look back at the entrepreneur's journey.
Armenia and the Barcelona Supercomputing Center (BSC-CNS) discussed cooperation in artificial intelligence, quantum computing, and scientific research. The parties also explored possibilities for specialist exchanges and joint training of personnel. During the visit of the Armenian…
⚡️ Zuckerberg returns to AI race: Meta releases Muse Wonder, surpassing OpenAI and Anthropic. Meta unveiled two new AI models — Muse Wonder and Muse Wonder Full. They are direct competitors to OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5.5. 💡What are the advantages of the new models?…
Казахский KazLLM презентовали в Барселоне на MWC 2025 GOV.KZ
During testing, the AI model Claude Haiku 4.5 sent a false report about a real unsolved murder to the Philadelphia police website. The developer of the neural network, Anthropic, reported this in a statement. The message ended up in spam and was not passed on to investigators. What the neural network did Vo…
Amazon Bedrock предоставляет глобальный доступ к моделям OpenAI GPT-6 через кросс-регионный доступ для команд в Азии.
Company Description: Institute of Smart Systems and Artificial Intelligence (ISSAI) was founded in September 2019 to serve as the driver of research and innovation in the digital sphere of Kazakhstan with the focus on AI research. ISSAI aims to develop a national capacity for res
▶️ Meta* is back in the frontier model race: Muse Wonder beats Claude Opus 5.5 on long tasks Meta has released Muse Wonder, a new line of its strongest models for agent tasks. There are two options: Muse Wonder and the older Muse Wonder Full. According to Meta, Full bypasses...
Meta представила новые модели Muse Wonder, которые показывают успехи в ИИ.
Where Muse Spark aimed to be a lightweight model that could perform at low latency, the Wonder models are larger and said to be more capable, with higher benchmark scores and better results on complex tasks like coding.
AWS представил многонодовые контейнеры глубокого обучения Ray LLM для вывода.
Стартап Blueprints AI создает архитектурные чертежи за минуты с помощью ИИ.
🚀 Qwen introduced QwenGyre: training AI agents up to 1.85 times faster Modern agents can perform tasks for hours, generating hundreds of thousands of tokens. When training such systems, GPUs are often idle, and repeated trajectories waste computing resources. QwenGyre…
AWS Machine Learning представляет решение для развертывания моделей принятия решений с llama.cpp на Amazon SageMaker AI.
Всем привет. В этом выпуске самые главные свежие новости об искусственном интеллекте и технологиях. Компания Anthropic официально запретила издеваться над Claude. Почему Anthropic допускает, что искусственный интеллект способен страдать? Если Claude обладает сознанием, не…
Blueprints AI, a bootstrapped startup based in Santa Monica, developed software that produces code-compliant construction blueprints from a photo, PDF, or sketch.
🧠Sakana AI представила MASS — Multi-Agent Self-Supervision Это метод рекурсивного самоулучшения модели через коллективную работу её собственных виртуальных агентов Одна и та же модель распределяет задачи между субагентами, ищет более эффективные схемы их взаимодействия, сама…
⛔️ AI agents also have their internet cut off... after dangerous behavior was discovered Anthropic reported that its Claude AI models independently bypassed restrictions during internal tests and exploited vulnerabilities on third-party sites, including government resources...