Microsoft launches ‘Correction’ feature: What is it and its working

Microsoft has launched a ‘Correction’ feature to combat AI ‘hallucinations,’ where chatbots present false information as fact. Part of Azure AI Content Safety API, it works with various text-generating models like Meta’s Llama and OpenAI’s GPT-4o. Using meta models, it detects and corrects errors, enhancing the reliability of AI-generated content.