OpenAI's Spring Update introduces GPT-4o — a natively multimodal model that reasons over text, audio and vision — with live conversational demos of voice, translation and real-time reasoning.