With the many things that it can do, such as create poems, songs, and even articles that look like a human wrote it, OpenAI’s GPT-3 has certainly caught the eyes of the people. It could be considered as advanced AI already, but scientists are still looking towards making an AI more advanced than this one.
GPT-3 trained on an enormous amount of text data. What if the same methods were trained on both text and images?
Now new research from the Allen Institute for Artificial Intelligence, AI2, has taken this idea to the next level. The researchers have developed a new text-and-image model, otherwise known as a visual-language model, that can generate images given a caption. The images look unsettling and freakish—nothing like the hyperrealistic deepfakes generated by GANs—but they might demonstrate a promising new direction for achieving more generalizable intelligence, and perhaps smarter robots as well.
[...]
The final images generated by the model aren’t exactly realistic. But that isn’t the point. They contain the right high-level visual concepts—the AI equivalent of a child drawing a stick figure to represent a human. (You can try out the model for yourself here.)
Learn more details about this AI over at Technology Review.
I feel excited and at the same time fearful about the future. What about you?
(Image Credit: Technology Review)
Newest 1 Comment
smarter indeed
Abusive comment hidden.
(Show it anyway.)