This avocado armchair could be the future of AI
OpenAI has extended GPT-3 with two new models that combine NLP with image recognition to give its AI a better understanding of everyday concepts.
✒ Imagine “an avocado armchair” and “an illustration of a baby daikon radish in a tutu walking a dog”. What comes to mind? Let’s see how artificial intelligence is able to visualise it for you.
This new text-to-image generator DALL·E — a smaller version of GPT-3 that has also been trained on text-image pairs taken from the internet — is seemingly able to generate synthetic images out of rather whimsical text.
Impressive? Uncanny? I don’t know how to feel.