Member-only story
OpenAI Just Killed Prompt Engineering With DALL-E 3, And Won’t Stop
I once asked ChatGPT if it could create pictures, and I got this reply:
I’m a text-based AI and cannot create pictures or visual content. However, I can help describe or suggest ideas for images if you’d like.
The chatbot’s ability to help describe or suggest ideas for images is what OpenAI used to make DALl-E 3, and it will dismantle prompt engineering for AI art as we know it.
The Big Deal With DALL-E 3
ChatGPT is for texts; DALL-E is for images. OpenAI owns them all.
All old DALL-E versions, like all AI text-to-image generators, had a shortcoming that made it very difficult for you to create a prompt that could translate your mental picture of something into the model.
Recently, however, OpenAI announced the release of DALL-E 3, which it said tackled this shortcoming.
In their own words:
DALL·E 3 understands significantly more nuance and detail than our previous systems, allowing you to easily translate your ideas into exceptionally accurate images.
Before now, getting good enough results from these text-to-image models was an art; it takes a good amount of articulation and an understanding of technicalities…
