Astana time
AI QAZAQSTANKZ · AIBeta

Smart portal about artificial intelligence and high technology in Kazakhstan

All stories

All published intelligence and news stories

7626 articles
Infrastructure
KZ·AI 339F
Основатель и CEO NVIDIA впервые в Астане: Дженсен Хуанг познакомился с AI-экосистемой Казахстана
Infrastructure🇰🇿Astana HubStory: Дженсен Хуанг, глава NVIDIA, выступит с речью в Астане 23 октября#Робототехника#OpenAI1 min

RUОснователь и CEO NVIDIA впервые в Астане: Дженсен Хуанг познакомился с AI-экосистемой Казахстана

23 октября 2026 года основатель и CEO NVIDIA Дженсен Хуанг впервые посетил международный центр искусственного интеллекта alem.ai в Астане. В ходе визита ему представили стратегию Казахстана по формированию глобального хаба Physical AI, основанную на трех ключевых составляющих:…

Министерство ИИ и цифрового развития РК (gov.kz)
Llama: Local AI on Your Mac
Knowledge base#Meta AI#YouTube1 min

Llama: Local AI on Your Mac

Llama runs AI models right on your Mac. It lives in your menu bar and suggests models that fit your Mac. One click to download, then open a chat, the model loads on its own. Your conversations never leave your computer and it even works offline. And any app that speaks the Open…

YouTube — Hugging Face
Everything Is Getting Hacked (it’s over)
Knowledge base#OpenAI#DeepSeek1 min

Everything Is Getting Hacked (it’s over)

My name is David Andre and AI powered hacking is here because we've just seen the most insane 48 hours in the history of cyber security. So first of all people are calling this the black October due to how many different vulnerabilities, exploits and hacks and also the severity…

YouTube — David Ondrej (AI Agents)
Startups
KZ·AI 3D9C
Elon Musk plans to merge Tesla and SpaceX - the potential value of the companies is $3.5 trillion
StartupsStory: Илон Маск задумал объединить электромобили, космос и ИИ в одном бизнесе#Робототехника#Илон Маск1 min

Elon Musk plans to merge Tesla and SpaceX - the potential value of the companies is $3.5 trillion

🚀 Elon Musk plans to merge Tesla and SpaceX - the potential value of the companies is $3.5 trillion Musk confirmed that he is considering the possibility of merging his two largest companies. According to him, the merger makes sense thanks to joint projects in the field of AI,...

Telegram — GPT Time | Нейросети | ИИ
Startups
KZ·AI 2CAC
NVIDIA invested in Kazakh startup Higgsfield AI
Startups🇰🇿 KazakhstanStory: NVIDIA инвестировала в казахстанский стартап Higgsfield AI#Nvidia#Дженсен Хуанг1 min

NVIDIA invested in Kazakh startup Higgsfield AI

⚡️️️️️️️NVIDIA invested in the Kazakh startup Higgsfield AI During a speech in Astana, NVIDIA founder and CEO Jensen Huang was asked why the company decided to invest in Higgsfield AI and what other startups should focus on in order to...

Telegram — BlueScreen.kz
Research
KZ·AI 2B82
Дженсен Хуанг: создание собственного ИИ — вопрос национальной безопасности Казахстана Основатель и генеральный директ...
Research🇰🇿KazLLM#Nvidia#KazLLM1 min

RUДженсен Хуанг: создание собственного ИИ — вопрос национальной безопасности Казахстана Основатель и генеральный директ...

Дженсен Хуанг: создание собственного ИИ — вопрос национальной безопасности Казахстана Основатель и генеральный директор NVIDIA Дженсен Хуанг во время встречи с министром искусственного интеллекта и цифрового развития Жасланом Мадиевым в Астане объяснил, почему…

Telegram — BlueScreen.kz
Day 06 | AI BUILDER SERIES | What is Microsoft Purview? #m365 #microsoft #copilot #cowork
Knowledge baseStory: День 05 | СЕРИЯ AI СТРОИТЕЛЬ | Ограждения в Microsoft Copilot #m365 #microsoft #copilot #cowork#Microsoft#YouTube1 min

Day 06 | AI BUILDER SERIES | What is Microsoft Purview? #m365 #microsoft #copilot #cowork

🔐 **Microsoft Purview Explained | Data Security & AI Governance** Discover how Microsoft Purview helps organizations protect sensitive data, manage compliance, apply data loss prevention, and govern data responsibly in the age of AI. Learn how Microsoft Purview supports secure…

YouTube — Microsoft Developer
Research
KZ·AI E724
ANTHROPIC
Research#Anthropic1 min

ANTHROPIC

ANTHROPIC: Starting next week, Claude will start using Auto Mode by default to manage permissions on Pro, Max and Team plans. In this mode, Claude will not ask users for confirmation of each action, but will instead rely on its own...

Telegram — Вайб-кодинг
Infrastructure
KZ·AI D8DA
3Blue1Brown style math animations are created using Manim. They look really impressive, however...
Infrastructure#Anthropic#GPU & Суперкомпьютеры1 min

3Blue1Brown style math animations are created using Manim. They look really impressive, however...

3Blue1Brown style math animations are created using Manim. They look really impressive, but rendering even one video takes quite a long time. ManimGX retains the familiar Python syntax of Manim Community, but uses Rust and GPU to...

GitHub Разработчика
Startups
KZ·AI CF7D
Kazakhstan entered the top 35 largest e-commerce markets in the world
Startups🇰🇿 Kazakhstan1 min

Kazakhstan entered the top 35 largest e-commerce markets in the world

🛒️️️️️️ Kazakhstan entered the top 35 largest e-commerce markets in the world. In 2025, online sales revenue in Kazakhstan reached $8.7 billion, according to ECDB. And the volume of retail electronic trade in goods, according to the Bureau of National Statistics, amounted to 3.77 trillion tenge - almost double...

Telegram — BlueScreen.kz
Policy & regulation
KZ·AI 5A0E
Digital Code of Kazakhstan: what to check in documents, digital signatures and personnel processes
Policy & regulation🇰🇿 Kazakhstan1 min

Digital Code of Kazakhstan: what to check in documents, digital signatures and personnel processes

🔽 Digital Code of Kazakhstan: what to check in documents, digital signatures and personnel processes The Digital Code is already in force. What does this change for business? What requirements need to be taken into account when working with electronic documents, how to properly organize their storage and what’s new in…

Telegram — BlueScreen.kz
Research
KZ·AI 770A
Chinese mathematician prepares 1,000 problems to challenge artificial intelligence
Research1 min

Chinese mathematician prepares 1,000 problems to challenge artificial intelligence

🧠 Chinese mathematician prepares 1,000 problems to challenge artificial intelligence. 61-year-old Nanjing University professor Sun Zhivei compiled a collection of 1,000 unsolved mathematical conjectures spanning over 400 pages. The scholar is confident that AI will not be able to solve…

Telegram — GPT Time | Нейросети | ИИ
Infrastructure
KZ·AI 9D87
Jensen Huang and Zhaslan Madiyev to Discuss AI Future: Join the Online Broadcast
Infrastructure🇰🇿 KazakhstanStory: Чат у камина: Дженсен Хуанг и Жаслан Мадиев о будущем искусственного интеллекта | Прямой эфир из Астаны#Nvidia#Дженсен Хуанг1 min

Jensen Huang and Zhaslan Madiyev to Discuss AI Future: Join the Online Broadcast

📱 Jensen Huang and Zhaslan Madiyev to Discuss AI Future: Join the Online Broadcast Today at 11:00 Astana time, NVIDIA founder and CEO Jensen Huang will meet with Deputy Prime Minister — Minister of Artificial Intelligence and Digital Development Zhaslan Madiyev…

Telegram — BlueScreen.kz
Research
KZ·AI 6540
Breaking the high-degree barrier in Lehmer's problem with a human-AI discovery loop
Research#arXiv1 min

Breaking the high-degree barrier in Lehmer's problem with a human-AI discovery loop

arXiv:2610.30277v1 Announce Type: new Abstract: While artificial intelligence has made rapid progress in mathematical reasoning, many open problems depend on structure that is revealed only through large-scale computational exploration and analytical synthesis. One such example…

arXiv — cs.AI
Research
KZ·AI C98C
Route-and-Generate: KL-Regularized Router Improvement for Reasoning Routing in LLMs
Research#arXiv1 min

Route-and-Generate: KL-Regularized Router Improvement for Reasoning Routing in LLMs

arXiv:2610.30351v1 Announce Type: new Abstract: Large language models often apply complex reasoning indiscriminately, leading to overthinking on simple tasks and underthinking on harder ones. We present Route-and-Generate (Rouge), a hierarchical training framework that enables…

arXiv — cs.AI
Research
KZ·AI 747E
Soft Ontology Priors over Latent Concept Posteriors for Weakly Supervised Neuro-Symbolic Learning
Research#arXiv1 min

Soft Ontology Priors over Latent Concept Posteriors for Weakly Supervised Neuro-Symbolic Learning

arXiv:2610.30391v1 Announce Type: new Abstract: Neuro-symbolic (NeSy) predictors that supervise only the task label suffer from reasoning shortcuts: distinct concept mappings achieve the same maximum likelihood, leaving the recovered concepts unidentifiable and unsafe for…

arXiv — cs.AI
Research
KZ·AI 8F40
From Single-Actor Environments to Multi-Actor Event Worlds: Can LLMs Be Event World Models?
Research#arXiv1 min

From Single-Actor Environments to Multi-Actor Event Worlds: Can LLMs Be Event World Models?

arXiv:2610.30488v1 Announce Type: new Abstract: World models help agents assess the consequences of their actions before acting. Many real systems, however, involve multiple actors who make their own decisions, can evolve even when the agent takes no action, and are only…

arXiv — cs.AI
Research
KZ·AI 835C
From Prose to Properties: Testing Research Claims with ARATest for Agent-Native Science
Research#arXiv1 min

From Prose to Properties: Testing Research Claims with ARATest for Agent-Native Science

arXiv:2610.30495v1 Announce Type: new Abstract: AI agents now run experiments and report findings, so research produces claims faster than researchers can check them. Each empirical claim implies a property that must hold over its cases, but it states that property only in…

arXiv — cs.AI
Startups
KZ·AI 9502
Governed Self-Evolution for Autonomous Investing: A Multi-Agent LLM System with an Invariant Core
Startups#arXiv1 min

Governed Self-Evolution for Autonomous Investing: A Multi-Agent LLM System with an Invariant Core

arXiv:2610.30557v1 Announce Type: new Abstract: LLM agents that act on live Web data must combine heterogeneous evidence, remember what they have done and improve from outcomes without weakening their own safeguards. We study this problem in autonomous investing, where every…

arXiv — cs.AI
Research
KZ·AI BE89
Large language models show systematic failures in incorporating evolving medical knowledge
Research#arXiv1 min

Large language models show systematic failures in incorporating evolving medical knowledge

arXiv:2610.30631v1 Announce Type: new Abstract: Medical knowledge evolves as clinical guidelines revise recommendations, yet whether large language models (LLMs) faithfully incorporate such changes remains unclear. We introduce ChronoEvo, comprising 1,709 recommendation-level…

arXiv — cs.AI
Research
KZ·AI A178
Better Reasoning, Not Just Bigger Models: A Structure-Grounded Framework for Efficient and Reliable Inference
Research#arXiv1 min

Better Reasoning, Not Just Bigger Models: A Structure-Grounded Framework for Efficient and Reliable Inference

arXiv:2610.30611v1 Announce Type: new Abstract: Scaling large language models has substantially improved linguistic fluency, yet reliable autonomous agents require more than fluent generation. In multi-turn inference, current agentic systems often produce semantically plausible…

arXiv — cs.AI
Research
KZ·AI 4828
The Cost of Saying Hello: An Empirical Study of Handshake Overhead in AI Agent Communication Protocols
Research#arXiv1 min

The Cost of Saying Hello: An Empirical Study of Handshake Overhead in AI Agent Communication Protocols

arXiv:2610.30151v1 Announce Type: new Abstract: Every interaction between AI agents begins with a handshake: the initiating agent must locate its counterpart, learn what it can do, and establish a session before the first useful message can be sent. Surveys of agent…

arXiv — cs.AI
Research
KZ·AI 2E03
MCP vs CLI for AI Agents: A Comparative Analysis of Tool Integration Strategies for Autonomous Systems
Research#arXiv1 min

MCP vs CLI for AI Agents: A Comparative Analysis of Tool Integration Strategies for Autonomous Systems

arXiv:2610.30157v1 Announce Type: new Abstract: AI agents require interfaces to interact with external systems such as files, APIs, databases, and cloud services. Two competing paradigms have emerged: CLI (Command Line Interface), where agents execute shell commands directly,…

arXiv — cs.AI
Research
KZ·AI B0ED
Same Beliefs, Different Policies: Learning Reusable Representations for Early Stopping in LLM Reasoning
Research#arXiv#GPU & Суперкомпьютеры1 min

Same Beliefs, Different Policies: Learning Reusable Representations for Early Stopping in LLM Reasoning

arXiv:2610.30450v1 Announce Type: new Abstract: Reasoning LLMs often continue reasoning after reaching a correct answer, wasting computation and sometimes replacing it with an incorrect one. Deciding when to stop is a metareasoning problem: the model must weigh what further…

arXiv — cs.AI
Research
KZ·AI A2B4
What the Trace Shows and What the Label Says: Auditing Behavioral Detectors in Agent Evaluation
Research#arXiv1 min

What the Trace Shows and What the Label Says: Auditing Behavioral Detectors in Agent Evaluation

arXiv:2610.30491v1 Announce Type: new Abstract: LLM agents are often evaluated by reducing execution traces to labels such as final-answer correctness, tool verification, and fault propagation. These labels are measurements produced by parsers, behavioral detectors, and harness…

arXiv — cs.AI
Research
KZ·AI 14E4
Accelerating Wildfire Hazard Mapping Across Canada with an AI Surrogate Modelling Framework
Research#arXiv1 min

Accelerating Wildfire Hazard Mapping Across Canada with an AI Surrogate Modelling Framework

arXiv:2610.30170v1 Announce Type: new Abstract: Canada's increasing wildfire activity requires frequent national hazard mapping to guide proactive mitigation, including fuel management, land-use planning, infrastructure protection, and evacuation routing. However, process-based…

arXiv — cs.AI
Research
KZ·AI A9C8
Feedback That Backfires in Games: Do LLM Players Repeat the Move They Just Watched Fail?
Research#arXiv#GPU & Суперкомпьютеры1 min

Feedback That Backfires in Games: Do LLM Players Repeat the Move They Just Watched Fail?

arXiv:2610.30205v1 Announce Type: new Abstract: When an LLM agent's move fails, the harness writes the move and the failure message into the transcript and asks again. In tool calling, this record makes small models more likely to repeat the failed action. We test whether the…

arXiv — cs.AI
Research
KZ·AI DF0E
SpecMAGIC: Specification-Guided Multi-Task Generalization in Multi-Agent Reinforcement Learning
Research#arXiv1 min

SpecMAGIC: Specification-Guided Multi-Task Generalization in Multi-Agent Reinforcement Learning

arXiv:2610.30443v1 Announce Type: new Abstract: Multi-task generalization is an important yet underexplored problem in multi-agent reinforcement learning (MARL). Many real-world applications require multi-agent systems to generalize across complex, temporally extended tasks with…

arXiv — cs.AI
Research
KZ·AI B106
Why Bother Me Again? Benchmarking Personal Agent Memory for Interaction Preferences
Research#arXiv1 min

Why Bother Me Again? Benchmarking Personal Agent Memory for Interaction Preferences

arXiv:2610.30485v1 Announce Type: new Abstract: Personal intelligence systems are taking on increasingly complex everyday tasks, yet repeated questions, unnecessary confirmations, and unwanted outreach can turn capable assistance into interaction burden. Reducing this burden…

arXiv — cs.AI