ChatGPT: Multimodal in the coming weeks
PLUS: Youtube new AI tools, Microsoft Copilot and more
Hi folks!👋🏻 This is The Prompt! We've got the screenplay for what's happening in the AI scene!
Lights, camera — Let's get it 🎬
ChatGPT+ will be multimodal in the coming weeks
OpenAI has announced (as speculated!) that they will embed their latest image model, Dalle-3, into ChatGPT, making it multimodal.
The key phrase in their announcement is that Dalle-3 comes as a ChatGPT upgrade.
What does this mean?
This means that Dalle-3 will have the GPT4 foundation making it much smarter and easier to use; something that’s not the case for other image models who require lots of prompt hacking.
And this setup might give this image model the one “skill” that we haven’t seen in other image models yet: generating text on images.
Winter is coming: Big Multimodal AI’s battle
Yes, this combo is a problem for Midjourney. But really, it just means we'll see a big fight between big, powerful multimodal AI’s this winter. Google’s Gemini model is rumored to be multimodal too.
Multimodal AI’s means that we’ll get to create images with very simple prompts. So, prompt engineering might be obsolete?! As things develop, that might be the case.
And the video they shared is nothing less than amazing.
🚨 What else is going on
Lexica introduced their latest next-gen image model, Aperture v3.5. More photorealistic, better prompt “following”.. yada yada - this kinda feels obsolete with the Dalle-3 news?
YouTube to add AI creator tools like AI language dubbing to AI video generation and ideation
Microsoft Copilot was announced yesterday. Copilot will incorporate the context and intelligence of the web, your data and what you are doing in the moment on your PC to provide better assistance
🧰 Tools of the trade
✍🏼 Prompt of the Day
"An illustration of an avocado sitting in therapist's chair, saying "I just feel so empty inside" with a pit-sized hole in its center. The therapist, a spoon, scribbles notes."