All published intelligence, sovereign stack research, ecosystem news, and AI foundation model reviews.
Compare GigaChat models on your task in GigaPlatform AI Studio This is an environment in which you can change the settings of the models and test the answers to your requests. In AI Studio you can: • Compare the responses of two models on one request. • Compare answers at different temperatures. • Configure…
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…
In a world of AI Overviews, Wikipedia usage declines. But the real loss may be far more existential than website traffic.
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…
Character.AI, one of the first breakout startups of the generative AI boom, has filed for bankruptcy and is selling some of its assets to Disney. The company, which builds various online bots for users to chat or roleplay with, has faced mounting scrutiny and lawsuits over the…
AI labs are starting to buy data from defunct startups to train their models.
🏗 AI Startup Promises to Create Construction Drawings in Minutes Blueprints AI is developing a system that turns a 2D sketch or PDF into a set of architectural documents, ready to be checked for compliance with building codes. In the new version it will be possible to immediately generate...
Backed by Silicon Valley titans, Leading the Future has raised huge sums to promote pro-AI candidates on both side of the aisle.
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…
MIT is exploring how AI influences accountability in civil discourse.
When Keelan Miskell began winding down his crypto startup this year, he didn’t think its data was worth much: the users were anonymous, the company just a few years old. But he’s since turned the firm’s digital archives into real money. He’s netted roughly $25,000 for his code…
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.…
Hello. And we continue the testing cycle for the Xiaomi model. They recently released the new MIM 2.6 Pro MIM 2.6 Flash. Mim 2.6 Flash is already on the channel, one of the previous videos. We will be reviewing MIM 2.6 Pro today. As I said in that video, I repeat with this model...
✔️ 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.
🧩 Artificial Analysis launches AA-3D. Models receive a text brief and must independently assemble an interactive three.js scene in an offline sandbox. The test comprised 55 tasks across seven categories, including architecture, vehicles, living organisms, games, and space. For each task, the model…
Итак, вы создаете приложение с помощью ИИ. Выглядит неплохо. Можно искать клиентов, открывать заявки в поддержку и нажимать кнопку для возврата средств. Работает отлично. Но потом вы передаете его кому-то еще в компании. Могут ли они оформлять любой возврат? Могут ли они просто…
When government officials craft policies, there is often a comment period for public feedback. Lawmakers have long had to sort through fake feedback that is mass-produced in an effort to sway policy, but they’ve generally gotten used to detecting it. With, artificial…
When I woke up this morning, I did not expect Meta to have a Frontier model. Yet, here we are. In fact, it's kind of beating every model on the planet. Maybe with the exception of Opus 5.5. Yeah, you heard me right. This is Muse Wonder and Muse Wonderful Full. Two models out of…
Paragon on air — wrapping up the day! Top stories for the day🟠12.10 🟠Meta — new AI model with memory 🟠Microsoft — call for stricter AI regulation 🟠iPhone Duo — pre-order for foldable iPhone from Oct 16 🟠Claude — monthly API credits 🟠OpenAI — more memory in GPT-6.1 Sol…
26 subscriptions to ChatGPT, Claude, and Gemini — giving away for free! If you missed it — reminder that we currently have two giveaways of top AI subscriptions. ▶️3 ChatGPT Plus + 10 Google AI Pro — results tomorrow, October 13, at 13:00 MSK ▶️3 Claude Pro + 10 Google AI Pro —…
⚡️ Microsoft is betting on its chips: Maia will become the company's main inference platform From 2027, the company will scale the Maia family to multi-gigawatt deployments, and Maia will become the main inference system for its own products and services. At the same time...
Amazon Bedrock предоставляет глобальный доступ к моделям OpenAI GPT-6 через кросс-регионный доступ для команд в Азии.
Meta's new Muse Wonder models rank in the top three on six out of ten benchmarks, with the biggest gains in agentic tasks. But the cost per task has nearly tripled. The article Meta's new frontier model Muse Wonder Full breaks into the AI top tier appeared first on The Decoder.
▶️ 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.
Google представила Fitbit Edge за $180 — со сменным аккумулятором 🔋 Новый фитнес-браслет заменит линейку Charge. Он получил овальный дисплей, GPS, круглосуточный мониторинг пульса, отслеживание температуры кожи и автоматическое распознавание активности. Аккумулятор работает до…
Sakana AI's MASS method lets one language model improve itself by solving tasks, judging results, and refining workflows, reaching a win rate of nearly 70 percent against the base model after two rounds. But when the model misses the same errors across all three roles, it can…
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.