A new AI model demonstrates significant improvements in processing complex, multi-step reasoning tasks by integrating a novel ‘chain-of-thought’ prompting technique. The system, developed by researchers at a leading AI lab, was tested on benchmarks requiring logical deduction and planning, outperforming previous state-of-the-art models by a notable margin. The approach focuses on training the model to break down problems into intermediate steps, mimicking a more human-like reasoning process. This advancement is seen as a key step toward developing AI capable of more reliable and transparent decision-making in fields like scientific research and logistics. Read the full article at https://technologyreview.com/2024/05/ai-reasoning-advance.