Users can interact with GPT-4o via text, audio, image and video, and get it to generate all of those forms of outputs. San Francisco-based OpenAI is offering some of its capabilities immediately, with others to be released in the near future. (The Logic)
Talking point: The firm launched the current era of generative AI buzz in November 2022 with ChatGPT, a question box that could reply to queries. But GPT-4o’s multimodal capabilities and faster response times make it closer to an always-on assistant, to whom users can assign tasks or show work for feedback. OpenAI is launching a desktop app for Apple’s Mac devices. It may soon migrate to mobile—the startup has reportedly agreed to a deal with the iPhone maker to integrate its AI into Apple’s next iOS iteration. OpenAI’s moves come the day before Google’s I/O developer conference, at which the tech giant has promised its own AI updates.