DALL·E and DALL·E 2, created by OpenAI, are machine learning models that have the ability to generate images from a short text description.
What is DALL-E?
|DALL-E and DALL-E 2, created by OpenAI, are machine learning models that have the ability to generate images from a short text description. DALL-E and DALL-E 2 are both forms of generative AI, creating images that may appear like real photographs or artwork created by a human. The main differentiating factor between the two systems are the number of parameters. DALL-E 2 can create higher-resolution images, using 12 billion parameters, versus DALL-E’s 3.5 billion parameters.
Other emerging text-to-image AIs are Midjourney and Stable Diffusion, both of which have their own unique approach to the process. Midjourney currently only runs as a bot via Discord (a popular chat platform) and tends to generate a more ‘painted’ look, whereas Stable Diffusion is open-source and has very few constraints on what can be generated.
While this opens amazing doors to the creative potential of machine learning, you should always be aware of those using the technology in bad faith. As AI generated images become more photorealistic, malicious parties may use this technology to forge images and information. Always be certain that the information you’re viewing has been secured and unaltered.