Any images that are produced by giving a text prompt to current generative AI models, such as Midjourney or Stable Diffusion, cannot be copyrighted in the US. That’s according to the US Copyright Office (USCO), which has equated such prompts to a buyer giving directions to a commissioned artist. “They identify what the prompter wishes to have depicted, but the machine determines how those instructions are implemented in its output,” the USCO wrote in new guidance it published to the Federal Register.
“When an AI technology receives solely a prompt from a human and produces complex written, visual, or musical works in response, the ‘traditional elements of authorship’ are determined and executed by the technology — not the human user,” the office stated.
It noted that the level of human creativity involved in a work is a significant consideration as to whether it will grant copyright protection. It suggested that current AI models can’t generate copyrightable work. “Based on the Office’s understanding of the generative AI technologies currently available, users do not exercise ultimate creative control over how such systems interpret prompts and generate material,” the USCO said. “In the Office’s view, it is well-established that copyright can protect only material that is the product of human creativity.” In one famous case, the office ruled that it couldn’t allow selfies taken by a monkey to be copyrighted.
When it comes to works that contain material generated by an AI, the USCO looks at whether the model’s contributions to the work are the result of “mechanical reproduction” (i.e., generated in response to text prompts) or if they represent the author’s “own mental conception.” Current rules state that the USCO “will not register works produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author.”
However, the office has left the door open to granting copyright protections to work with AI-generated elements. “The answer will depend on the circumstances, particularly how the AI tool operates and how it was used to create the final work,” it said. “This is necessarily a case-by-case inquiry. If a work’s traditional elements of authorship were produced by a machine, the work lacks human authorship and the Office will not register it.”
Last month, the USCO determined that images generated by Midjourney and used in a graphic novel were not copyrightable. However, it said the text and layout of Kris Kashtanova’s Zarya of the Dawn could be afforded copyright protection. The office said there was too much “distance” between Kashtanova’s inputs and Midjourney’s output for the images to be copyrightable. Kashtanova’s lawyers have said that by focusing on the output rather than the input, the office “applied the wrong legal standard.”
Meanwhile, the USCO has started an initiative to further explore copyright law and policy issues related to AI following requests from Congress and the public. It will host several panel discussions on the topics in April and May. The office plans to solicit public comments later this year on a swathe of copyright issues relating to the use of AI.
This article originally appeared on Engadget at https://www.engadget.com/ai-generated-images-from-text-cant-be-copyrighted-us-government-rules-174243933.html?src=rss
Brought to you by USA Today Read the rest of the article here.