Your art is controlled by a simple text prompt. Prompts can be very long or very short, but they are always capped at 400 characters.
When you add text to your prompt, this text is sent to the AI engine, which in turn generates an image based on your command.
Your prompt should be as detailed as possible. You can specify the art style, background details, lighting, angles and even camera styles to your prompt to increase the specificity.
A prompt like "a clown" will produce very different results from a prompt such as "an angry clown searching for money in a post apocalyptic world". We have demonstrated this below.
OpenAI has studied over 650 million images and captions. There is no definitive "guide" on how to generate the best art. We have to experiment and find styles which we enjoy!