If you're pretty familiar with creating chatbots in Botpress, creating GPTs, and building Assistants in the OpenAI playground, I'd love for you to help clarify some things for me. I've been out here in the streets taking it all in, watching tutorials on the new GPTs and other things, learning their limitations and benefits etc. and I'm just trying to make sense of it all. If we want to create GPTs and by this, I mean using the GPT Builder where there is also access to web browsing and DALL E etc...we can ONLY use these features and functions within the ChatGPT UI and ONLY people with a Plus account can access it if you were to share it publicly or via a link. You can add Actions but you're limited to ONE API root URL but you can have multiple endpoints OR use AI Actions by Zapier and be able to connect it to a ton of apps with a ton of actions but still, only those with a ChatGPT Plus account can use it. There's no way of putting this on a website or any other channel. If we want to create "GPTs" that we can integrate with a Botpress workflow conversation so we can put it on a website or other channel, we have to use the OpenAI playground to first build an "Assistant" and use the API Key and Asst ID then some coding in Botpress to connect it. The thing is from what I've understood, the Assistant doesn't have web browsing or can use DALL E like a "GPT". Though, you could add a ton of functions/actions to it by connecting with multiple external software to do advanced actions compared to a "GPT". I've seen an example where they've hosted the Assistant on Replit instead of building it in the Playground and there's a ton of coding that has to be done but you can connect it to Botpress this way as well but you need to really be tech-savvy for this one. From what I've seen, they all have their benefits and limitations and it comes down to your SPECIFIC use case this will us know which route we want to go with. Looking for clarity