Artificial intelligence is changing the way we perceive reality. In 2021 alone, AI researchers have produced a wealth of deep-learning models capable of generating entirely new images from simple text cues, expanding the human imagination.
Researchers at initiatives from OpenAI, Google, Facebook and other tech giants have developed text-to-image tools that have not yet been released to the public, but despite all this, even beyond the alpha phases of said AI models, we have the opportunity to experience the wonders of text-to-image AI. For example, proprietary solutions have proliferated in the open source space and among smaller companies like Midjourney. VOX tested one of these tools in this impressive video:
These AI tools represent a huge cultural shift because they take the technical work out of the image creation process. Instead, they rely on creative ideas, skillful use of language, and user taste. We’re still a long way from a final solution, of course, but what Midjourney, OpenAI and co are already demonstrating gives us hope for a revolution in image creation and could spell the end of image databases.
Post image: Midjourney