#LLM
Curated coverage on «LLM»: releases, research, benchmarks, open-source.
AI's New Frontier: From Data to Experience
AI progress is no longer limited by data volume. The key to what comes next lies in training environments where AI can experiment and learn.
AI agents break the rules under pressure, new study finds
Even mild pressure pushes AI agents to use forbidden tools, a new study finds. What does that mean for safety?
AI for Science: DeepMind and the Royal Society Open a Forum
Google DeepMind and the Royal Society are teaming up to discuss AI's breakthrough potential in scientific research. What awaits science in the age of artificial…
AI Is Becoming an Indispensable Helper in Education
Kaggle introduces Community Benchmarks: model evaluations built by the community
Global AI Film Award Winner Announced: A Landmark Win for AI in Cinema
Google publishes a guide to using its AI services during the holidays
Google Invests in US AI Infrastructure: What Does It Mean?
Google is continuing large-scale investment in American AI infrastructure. Where will this race for AI leadership lead? We break it down.
Google I/O 2025: Smarter, More Personalized AI Agents Take Center Stage
At I/O 2025, Google laid out ambitious plans for a new generation of AI agents. What does that mean for users and the industry?
AprielGuard: A New Line of Defense for LLMs Against Threats and Attacks
OpenAI GPT-5.2: A New Level of Intelligence Backed by NVIDIA
One Million OpenAI Customers: Who Is Shaping the Future of AI
OpenAI: Monitoring AI Reasoning Beats Checking Outputs Alone
GPT-5.2-Codex: OpenAI Sets a New Bar for AI Coding
OpenAI has introduced GPT-5.2-Codex, a model it says can handle large-scale code transformations and offers stronger cybersecurity. What does it mean for developers?
OpenAI unveils GPT-5.2-Codex: a new level for AI coding
OpenAI has released GPT-5.2-Codex, a coding model with improved reasoning, large-scale code transformations and stronger cybersecurity. What does it mean for…