OpenAI's latest experiment just dropped called DALL·E, a sort of GPT-3 for images. DALL·E is a trained neural network that creates images from text captions.

The adorable name is a portmanteau of the artist Salvador Dali and Pixar's WALL-E. A fun nod to the surreal artistic output the AI can create through mashups of textual prompts.

DALL·E is described by OpenAI as being capable of a diverse set of outputs including anthropomorphising animals and objects, as illustrated with the cute dog-owning daikon below:


DALL·E can also combine unrelated objects to create real-looking outputs like these avocado chairs. Some of which wouldn't look out of place in a startup office IMO.


These are just a couple of examples OpenAI has shared about DALL·E so far, with real-world applications still somewhat unknown.

You'll be pleased to hear we're not replacing our human design team with AI.

As with GPT-3 we're intrigued to see what will happen next with DALL·E. Since its launch in July, we've seen tons of interesting products launch leveraging GPT-3.

We can't wait to see what you can build utilizing DALL·E.

The cinema may still be a no-go for many of us right now but that doesn't mean we can't still watch our favorite films together. Hearo launched today and will let you safely watch your shows with friends and family, from the safety of your own homes.

Don't forget to order in your own popcorn. 🍿🎬