AI

Designers draw, developers code, but a gap still exists between them. Locofy introduces an intelligent "agentic layer" that bridges this divide and turns Figma designs into clean code in moments.

Companies are pouring billions into AI agents for project management, yet real productivity gains remain elusive. Is the problem poor-quality software, or is it the state of your data? Find out why off-the-shelf tools just aren't enough.

Anthropic's Claude Fable 5 model outperformed the competition in tests measuring the ability to solve real-world software tasks. With an 87.5% success rate on the Ramp SWE-Bench benchmark, it defines a new frontier for autonomous AI agents.

OpenAI introduces LifeSciBench — 750 real-world research tasks where even the best AI model barely passes a third. Seven scientific domains, 19,000 criteria. A cold shower for anyone who thinks AI is about to replace scientists.

Chinese humanoid robot maker AgiBot made its grand European debut at VivaTech in Paris, showcasing dancing robots and manipulation demos while confirming that the European market is a top priority.

Epic Games has unveiled its plans for Unreal Engine 6. The integration of advanced Claude and Gemini models directly into the development environment promises to accelerate programming and error analysis, opening up new possibilities for studios worldwide and in the Czech Republic.

The traditional approach to autonomous vehicle safety relies on miles driven. American startup Kodiak is taking a different path — combining safety methodology from nuclear power plants with an AI-powered edge-case hunter. The result is a system that can tell you not just how big the risk is, but also how certain it is about its estimate.

OpenAI is losing its absolute dominance. While ChatGPT still leads in user numbers, its market share has dropped below 50% for the first time. Why are people switching to Google Gemini and Anthropic Claude?

When Washington ordered Anthropic to cut off foreigners from the Claude Fable 5 and Mythos 5 models, it confirmed the darkest fears of European strategists. A single political decision is all it takes for Europe to lose access to cutting-edge artificial intelligence.

The GLM-5.2 model from China's Zhipu Z.ai represents strong competition for GPT-5.5. With a 1-million-token context window and open weights, it's changing the game for developers and businesses seeking efficiency.