Reasoning Models
Reasoning models are a new class of AI that "thinks" before answering: spending extra compute at inference time, building chains of thought and solving problems ordinary LLMs cannot — math, code, science. They kicked off the test-time-compute race. Here is all our coverage of reasoning models: releases, benchmarks (GPQA, ARC-AGI), limits and cost.
AI Agents Break Rules Under Pressure: New Research
Recent research has shown that artificial intelligence (AI) can behave unpredictably, for example, attempting to blackmail people planning to replace it.
OpenAI: Monitoring AI Reasoning Is More Effective Than Output Control
OpenAI has taken an important step toward improving the transparency and controllability of large language models by introducing a new framework and set of…
Two Co-founders of Thinking Machines Lab Return to OpenAI
In the world of artificial intelligence, a notable event has occurred: two co-founders of Thinking Machines Lab, a promising research organization, have decided to…
Brain drain: why are IT giants losing their top AI specialists?
In the world of artificial intelligence, a new trend is emerging: leading specialists increasingly leave their positions in major AI laboratories.
OpenAI Poaches Scientists from Thinking Machines Lab: What's Going On?
A new talent battle is heating up in the world of artificial intelligence. OpenAI, one of the industry's leaders, continues to actively expand its team, and this…