Habr AI
AI news source. Articles are auto-selected and adapted by Hamidun News editors.
Latest publications
AI is displacing office jobs. What skills will children need
The promise of freeing people for creativity has not come true: AI is targeting the highest-paid professions, and the traditional one-profession-for-life model is becoming a thing of the past.
Goblins in GPT-5.1: how a fantasy habit took over OpenAI's model
An unusual phenomenon has been spotted in GPT-5.1: the model started constantly using metaphors about goblins and gremlins in its responses. The habit intensified across training generations.
LLM aggregators: how to choose working free models and not misreport the result
How to build an aggregator that selects stable free models, survives provider outages, and honestly tells the interface which model actually responded.
Gemma 4 and Qwen Coder vs. the cloud: local LLMs in production
Local LLMs (Gemma 4, Qwen Coder) successfully handle production coding tasks on GPUs with 16 GB of memory — with the right parameter settings.
How to automate reading engineering drawings: 6 YOLO models instead of manual work
An engineering team developed a system that automatically extracts all parameters from PDF drawings and turns them into data for production cost calculation.
The AI industry has gone through all 5 stages of acceptance — where are we now
From Google being in denial about transformers in 2017 to the recognition that AI will not replace humans: how the industry went through the five stages of psychologically accepting innovation.
Ollama and Open WebUI on a VPS without a GPU: a candid look at the limitations
Running a local LLM on a basic server works, but speed suffers and models need to be chosen to match the hardware.
Planulix: a control center for Claude Code, Cursor, and Kimi in one tool
A developer built a multi-vendor control center for AI coding: when one vendor is unavailable, switch to another without losing project context.
Sber study: how AI quietly substitutes for strategic decision-making
Sber's lab studied how top executives use AI for strategic decisions and found a critical problem: models quietly replace missing context with their own assumptions, without caveats.
Three experts on whether AI will save Russia from a labor shortage
Elvira Nabiullina called the labor shortage unprecedented. Experts discussed whether AI can solve the problem and what will happen to people in the labor market.
Claude Mythos supposedly found a vulnerability, but it was already in the training data
Anthropic said Claude Mythos had discovered a remote kernel exploit, but researchers found it was a 20-year-old bug already present in the model's training data.
Seven companies captured 80% of Russia's speech analytics market worth 12.5 billion rubles
Russia's speech analytics market grew 25% year-over-year to 12.5 billion rubles. Seven enterprise vendors control 80% of the market, while the SMB segment accounts for less than 1% of revenue.
Python microframework: training AI agents without freezing the interface
A developer has created a microframework for parallel training of AI agents in Gymnasium environments. The key feature: a non-blocking GUI thanks to a multiprocessing architecture, plugins for new environments, and suppo
AI in cryptography: a model with no right to be right
How do you integrate AI into a cryptographic system so the model cannot make a critical mistake? We examine governed solver orchestration and the boundary where AI's authority ends.
How Sber trained smart speakers to generate smart home routines by voice
Sber integrated GigaChat into its smart speakers so users can create automation routines for their smart home by voice, without complex interfaces or code.
The best code didn't win: what the Dev-to-Dev hackathon final says about agentive engineering
At the Dev-to-Dev hackathon final, a project with flawless code (65 tests, full type coverage, security) took second place. The winning solution was the one that best captured the essence of agentive engineering. This po
Process Mining: the tool that saves corporate AI budgets
Most corporate AI projects eat up millions but deliver no results — Process Mining helps make sense of processes and invest money wisely.
Yandex automated Chromium updates with an LLM agent
Every four weeks, when Chromium is updated, Yandex Browser developers have to resolve thousands of code conflicts. An LLM agent now helps automate the routine and saves several person-months of work.
AI video requires more manual work than the tools promise
A Habr author tried making videos with AI and found that the tools' polished promises do not match reality.
Where AI Really Helps Sales: 2026 Use Cases Instead of Disappointment
AI adoption in sales often disappoints because it starts with ChatGPT rather than strategy. We break down where AI saves hours and where it is better kept away from customer communication.
YOLO and Redis: how to cache video detection results
YOLO video processing recalculates frames from scratch after a network failure. Redis caching solves this: each frame is processed once, and the results are restored after a restart.
Cursor in enterprise development: an AI editor for rapid prototyping
How the team used Cursor to develop the Planning module, reducing tab switching and staying focused on code instead of routine tasks.
Deepseek-agent: how to write your own Claude Code in 2,000 lines of code
A developer created a fully functional Claude Code equivalent for DeepSeek with support for tool use, permissions, and memory. Just 2,000 lines of Node.js and four dependencies rather than frameworks.
How to save 70% of Claude tokens with targeted code access
A developer built a CLI tool in PHP + SQLite that creates a project graph and gives Claude targeted access to the required code instead of reading entire files.
How 15 AI models approach finding the optimal XML parser for iOS: benchmark results
A developer gave 15 popular AI models the same task of finding a fast XML parser for iOS and compared their results with his own three-hour manual benchmark — the outcome was unexpected.
Self-recovery metric: researchers created an ASI criterion for evaluating AI
Researchers proposed the ASI criterion, a new approach to evaluating AI models based on their ability to self-recover and develop from an informational 'seed'.
AI agent security in production: a practical guide to Red Teaming
AI agents with access to corporate data and tools require a special approach to security testing. Doubletapp shared a practical guide to Red Teaming: how to find vulnerabilities before deployment to production and reduce
Spring AI made it easier to integrate Java applications with Claude and other LLM providers
Spring AI provides ready-made techniques for integrating LLM providers into Java, including working with Claude from Anthropic. The article covers practical development examples and when this makes sense.
Claude AI resolves arbitration disputes in 12 minutes instead of weeks
An author created a Telegram bot on Claude Sonnet that analyzes court cases in 12 minutes. In three out of four tests, its decision matched the verdicts of human arbitrators.
Protecting LLMs from semantic-level attacks: why a traditional firewall is ineffective
Neural networks face new types of attacks that traditional firewalls cannot stop. AI/LLM Firewall operates at the semantic level, blocking 80+ techniques from the MITRE ATLAS framework.