Users are frustrated with ChatGPT's lack of context when performing tasks like generating presentations from documents. The AI needs to understand personal context (user, audience, goals, work environment) to produce relevant and high-quality output, moving towards an 'ambient or personal AI' that 'follows' the user.
One reason that people get frustrated with generative AI (ChatGPT) is that it lacks context. But this is about to change. Let's say that I want to turn a document into a PowerPoint presentation. Today, if I give ChatGPT a simple prompt ("Read the attached document and turn it into a presentation") it is unlikely to do a good job because: => It does not know much about me => It does not know much about my audience => It does not know much about my goals => It does not know whether this is a work presentation or for fun. And if it is a work presentation, it does not know where I work, who my colleagues are, etc. And on and on. But this will soon change because we will have ambient or personal AI. This AI will "follow" you everywhere: everything that you see and read, it will see and read; everything that you hear, it will hear; everything that you type, it will remember. So then the AI will have context and you will be able to work with it just like you do with a colleague. Except that this personal AI will cost 1% as much, never get tired, and have vast intelligence. And it won't lie to you or play politics. (image: Gemini)