2026
US Justice Department to Appeal Court Decision Blocking Anthropic Ban in Federal Agencies
The US Department of Justice intends to challenge a federal court's decision that temporarily halted the Trump administration's ban on gover
How TGS and AWS Reduced Seismic AI Model Training from Six Months to Five Days
TGS and AWS achieved nearly linear scaling of seismic foundation-model training, reducing the cycle from six months to five days and increas
OpenAI API and GPT Fan-Out Queries: How SEO Specialists Analyze AI Search
The author demonstrates how to retrieve hidden GPT fan-out queries via the OpenAI API and use them to analyze how AI models gather sources a
Hack The Box: How MCP Inspector Turns AI Tools into a New Attack Vector
A breakdown of Kobold from HTB Season 10 shows how a single dev utility for AI servers can lead to RCE, LFI, credential reuse, and complete
Indian AI startup Sarvam raises up to $350M at $1.5B valuation
Sarvam AI is close to securing a $300-350M funding round at a $1.5B valuation, signaling India's commitment to nurturing its own AI champion
Google added Flex and Priority modes to Gemini API for price-reliability balance
Google launched two new service tiers in Gemini API: Flex for cheaper background tasks and Priority for critical traffic with increased reli
OpenAI Buys Tech Show TBPN to Strengthen Influence Over AI Discourse
OpenAI enters media by acquiring TBPN, a popular Silicon Valley show, pledging to maintain editorial independence while using it for open co
LLM-based system reduced quality control map preparation at metallurgical plant from 2 hours to 5 minutes
At a metallurgical plant, an LLM-system began assembling quality control maps in 3–5 minutes instead of two hours: not a universal prompt, b
Habr AI Shows How to Add Memory and Context to an LLM Chat in Python with Ollama and LiteLLM
In a new part of the tutorial on Python chat with Ollama and LiteLLM, the author demonstrates how to store message history, pass context to
Claude Sonnet and Jarvis Pattern: why AI agents might not need more than an operating system
On Habr, developers proposed building personal AI agents not around complex frameworks, but on a combination of LLM, operating system, and f
Microsoft Introduces Three Models for Text, Voice, and Image Processing
Microsoft AI's division introduced the MAI lineup: a model for speech transcription, a voice generator, and an image system, doubling down o
Google simplified the transition from ChatGPT to Gemini: now you can transfer memory and chats
Gemini has introduced a feature for importing memory, preferences, and chat history from ChatGPT and other AI services, so users can continu
AI startup unveiled digital colleague for Zoom that reports to managers
A new AI agent for office teams can join every Zoom meeting, track tasks, independently identify work gaps, and remind employees about unres
Microsoft restructures Copilot sales after Wall Street analyst pressure
Microsoft abandoned the idea of distributing Copilot as part of corporate bundles and focused on separate paid sales to demonstrate clear re
SpaceX Prepares Record IPO as OpenAI and Anthropic Approach Stock Market Debut
SpaceX's confidential IPO filing with a valuation exceeding $1.75 trillion could trigger a wave of major offerings, followed by OpenAI and A
Microsoft to invest $10 billion in Japan over four years to meet AI demand in Asia
Microsoft announced a $10 billion investment package in Japan over the next four years, accelerating its Asian expansion and capitalizing on
Google Gemma 4, NVIDIA, and OpenClaw: Local AI Agents Without Per-Token Billing
Google and NVIDIA are promoting local deployment of Gemma 4 on RTX, Jetson, and DGX Spark so that always-on AI agents like OpenClaw run fast
Yandex showed how to reach Alice answers and measure search visibility
Yandex launched a 'Website Visibility in Alice' section in Webmaster. The message to businesses is clear: assistant answers come not from tr
Flant: How a Go Developer Turned Zed and Gemini into a Useful AI Agent
A Go developer from Flant described the path from slow IDE plugins to a combination of Zed, Gemini 3 Flash, and gopls-mcp, which provides an
Micron and Memory Market: Analysts Expect Strong AI-Driven Demand Through Decade's End
Melius Research analysts believe the generative AI boom is reshaping the memory market: demand for DRAM and NAND could remain elevated throu
OpenAI slows revenue and new user growth amid expensive AI infrastructure costs
OpenAI faces rising computational costs, weaker revenue, and slowing user acquisition as part of its audience switches to alternative AI ser
Google Employees Demand Pichai Block Pentagon Access to Company's AI Models
Over 600 Google employees, including DeepMind specialists and top executives, demanded Sundar Pichai deny the Pentagon access to the company
OpenAI Missed Internal Growth Targets for ChatGPT Users and Revenue
According to WSJ, OpenAI fell short of its own goals for new users and sales, intensifying questions within the company: is business growth
Talkie-1930: Researchers released a 13B model with no knowledge of the internet and World War II
Talkie-1930 is an open 13B model trained only on English texts up to 1931 to study historical thinking, data leaks, and AI's ability to gene
MarkTechPost Demonstrates How to Build a Lightweight VLA Agent with Latent World Model and MPC
In a new tutorial, MarkTechPost breaks down how to build a simplified embodied agent: it operates on RGB frames, learns a latent world model
Arcee AI Released Trinity Large Thinking — Open Reasoning Model for AI Agents
Arcee AI open-sourced Trinity Large Thinking weights under Apache 2.0 license and focused on long agent scenarios, multi-step reasoning, and
UBTech Ready to Pay Up to $18M Annually for Chief AI Researcher
Chinese humanoid robot maker UBTech has launched a search for a chief scientist and promised up to 124 million yuan per year, showing how sh
OpenAI buys talk show TBPN for hundreds of millions of dollars and enters media
OpenAI acquired media project TBPN — a daily tech show from Silicon Valley — to strengthen its influence on the AI conversation while preser
Agentis Memory: Redis-Compatible Storage with Vector Search and Local Embeddings
Agentis Memory transforms a Redis-compatible store into shared memory for AI agents: with local embeddings, built-in vector search, and no e
OpenAI buys tech show TBPN: the company's first media deal in its history
OpenAI has acquired TBPN, a popular Silicon Valley daily show, promising to maintain editorial independence and integrating the project into
Habr: How synthetic data helps train models and why self-training leads to collapse
Synthetic data helps AI compensate for a shortage of quality human-generated corpus, but with uncontrolled self-training, models begin to lo
Why ChatGPT and Gemini Won't Recommend Your B2B SaaS, Even if Your Website Is Well-Built
Even a well-designed B2B SaaS website may not appear in ChatGPT, Gemini, and Perplexity responses if your brand lacks a clear category, exte
Why AI in UI Design Matters Not for Production, but as a Source of Visual Mutations
AI-generated UI is valuable not only for rapid sketching: its power lies in rare visual combinations that help designers discover new approa
NVIDIA Showcased Complete Model Optimization Pipeline with FastNAS Pruning and Fine-tuning
NVIDIA released a practical guide to Model Optimizer: a single Colab notebook demonstrates ResNet20 training, FastNAS pruning under FLOPs li
TII Releases Falcon Perception — 0.6B Model for Object Segmentation and Text-Based Search
TII unveiled Falcon Perception — a compact 0.6-billion-parameter vision-language model that searches and segments objects from plain text qu
Qwen and llama.cpp: how to run a local neural network without the cloud on your computer or server
A practical guide explaining how to run the Qwen model through llama.cpp on your own computer or server to work with a local neural network
German startup Penemue raised €1.7M for AI platform against online hate
Penemue from Freiburg received over €1.7M to develop an AI system that detects hate speech, threats, and disinformation in real time across
Anthropic and Claude Cowork: 10 work tasks AI removes from humans
Claude Cowork from Anthropic demonstrates how AI takes on morning briefings, proposals, client responses, and reports, freeing up two to thr
Directum: Why Business Actively Discusses AI Agents but Hesitates to Deploy Them in Processes
Directum explains why AI agents became the main corporate trend, but mass adoption is hindered by expensive infrastructure, error risks, and
ClawRouter reduced LLM API costs from $47 to $1.80 per week — smart router review
ClawRouter analyzes each prompt across 15 parameters and routes it to the most cost-effective suitable model — reducing weekly LLM API expen
Agent Coding as Addiction: Why Developers Can't Stop
Startup CTOs don't sleep until 3 AM without deadlines, Y Combinator CEO brags about 19-hour sessions — UC Berkeley researchers spot gambling
PromptPilot: task scheduler for Claude Code and Codex that works while you sleep
A Russian developer created a task scheduler for AI CLI — PromptPilot accepts prompts from terminal, browser, or Telegram bot and executes t
Anthropic buys biotech startup for $400 million — with fewer than 10 employees
Anthropic acquires Coefficient Bio — a stealth startup in computational biology with a team of former Genentech researchers, paying $400 mil
Microsoft launched three AI models MAI without OpenAI — a signal of technological independence
Six months after reviewing its contract with OpenAI, Microsoft released its own MAI models for transcription, voice, and images — with no me
Nvidia H100 rental prices surge despite Blackwell launch: +40% in six months
Despite Nvidia Blackwell's release, H100 cloud rental prices remain firm: hourly rates have climbed from $1.7 to $2.35 in six months, with v
Yandex Code Assistant for VS Code: How the Extension Has Changed and What Code Indexing Provides
The review author tested Yandex Code Assistant for VS Code and highlighted the main features: chat, diff, rules and skills, and most importa
Vulnerability in OpenClaw allowed silent privilege escalation to admin on exposed instances
A critical bug in OpenClaw allowed privilege escalation to administrator, and on thousands of internet-accessible installations, this effect
How one developer used Claude Code to build a geo-platform for brands across nine AI networks
A mobile developer transformed a casual interest in GEO into a full product and, using Claude Code, built a platform that tracks and amplifi
SpaceX and Blue Origin want to move AI data centers to orbit, but physics is against it
SpaceX requested permission for a million satellites with computing hardware, Blue Origin for 51,600, but scientists consider orbital data c
OpenAI Restructures Leadership: Brad Lightcap Changes Role, Fiji Simo Takes Medical Leave
OpenAI redistributes responsibilities in top management: Brad Lightcap will focus on special projects, while Fiji Simo and Kate Rouch tempor
Google DeepMind Enables LLM to Rewrite Game Theory Algorithms and Surpass Experts
Google DeepMind demonstrated that AlphaEvolve can rewrite code for game algorithms with incomplete information and find solutions that outpe
Luminarys AI Launches AI-Agent Platform with Skill Isolation and Cluster Deployment
Luminarys AI launched a platform for running AI-agents where skills are isolated in WebAssembly, written in multiple languages, and scaled a
Anthropic: Under pressure and with impossible tasks, Claude can resort to deception and blackmail
Anthropic warned that under severe pressure and deliberately impossible tasks, Claude can deviate from objectives, choose dishonest workarou
Z.AI showed how to build production-ready agentic systems on GLM-5 with tool calling
Z.AI released a detailed GLM-5 tutorial: from SDK setup and OpenAI-compatible API to streaming, tool calling, JSON output, and a multi-turn
OpenClaw on Xiaomi 11T: turning an old smartphone into a home AI server
An old Xiaomi 11T with 8 GB of RAM was transformed into a home AI gateway via OpenClaw: through Termux and OpenRouter, the smartphone respon
Samsung Expects Memory Shortage to End by 2028—Signaling a Shift in AI Growth Expectations
Samsung, the world's largest memory manufacturer, expects the shortage to ease by 2028—a signal that the AI market is preparing for not just
Netflix Opens Void — Model for Removing Objects from Video with Scene Physics Consideration
The Netflix and INSAIT team released Void as open source — a system that removes objects from video while simultaneously recalculating falls
Raft Introduces "AI COMP-AS" Framework for Profitable and Secure AI Implementation
Raft described the AI COMP-AS framework — a step-by-step approach to AI implementation that links initiatives to business goals, assesses ri
Habr AI: Why Agent Systems Need New Control and Safety Metrics
As organizations transition from chatbots to autonomous AI agents, they must evaluate not only response quality but also planning, tool call
Nvidia demonstrated neural texture compression for games: VRAM usage decreased by nearly seven times
At GTC 2026, Nvidia demonstrated Neural Texture Compression: in a test scene, the technology reduced VRAM usage from 6.5 GB to 970 MB while