#LLM
Curated coverage on «LLM»: releases, research, benchmarks, open-source.

Google stopped the first zero-day developed with AI
For the first time, Google detected a zero-day vulnerability created with AI — cybercriminals were preparing it for a large-scale attack on

OpenAI launched Daybreak for automated vulnerability discovery in code
OpenAI launched Daybreak, an initiative for automatically detecting and fixing vulnerabilities in code before hackers find them.

Google adds Gemini Intelligence to Android: agents and smart widgets

Google unveils AI laptops and agentic Gemini features ahead of I/O

Google adds Gemini dictation to Gboard, posing a threat to startups

Asexual users find intimacy in conversations with AI assistants

Sasha Luccioni: sustainable AI is impossible without data on emissions and real-world usage
Researcher Sasha Luccioni believes the AI industry lacks basic transparency: companies rarely disclose energy use, emissions, and which mode

Clawdmeter turns Claude Code stats into a compact desktop panel
The open-source project Clawdmeter shows Claude Code usage stats in a small desktop panel, so developers can track activity without unnecess













