The massive change from GPT-3.5 is OpenAI's 4th technology language product is multimodal, which implies it can system both textual content, photos and audio. This suggests you may demonstrate it photographs and it will respond to them together with a text prompt – an early example of