The US Copyright Office States Images Created by Generative AI Models Cannot Be Copyrighted
To improve your local-language experience, sometimes we employ an auto-translation plugin. Please note auto-translation may not be accurate, so read original article for precise information.
The US Copyright Office has classified Midjourney and Stable Diffusion as generative AI models that cannot be copyrighted in the US.
The USCO believes prompts given AI models get are similar to a buyer’s directions to an artist.
The guidance states that when an AI technology receives solely a prompt from a human, it produces complex written, visual, or musical works in response.
The US Copyright Office (USCO) has recently stated that any images generated by existing generative AI models like Midjourney or Stable Diffusion cannot be protected by copyright laws in the US, according to new guidance published in the Federal Register. The USCO explains that the prompts provided to these models are similar to directions given by a buyer to an artist, where the machine determines how to implement these instructions in its output.
The USCO has highlighted the importance of human creativity in determining whether a work can be granted copyright protection. It stated that current AI models are not capable of generating copyrightable work because users do not have complete control over how the systems interpret prompts and generate material. The USCO firmly believes that copyright can only protect material that is the result of human creativity: The office once refused to allow a monkey’s selfies to be copyrighted.
The US Copyright Office has some interesting rules when it comes to works that have been created with the help of artificial intelligence. The USCO examines whether the AI model’s contribution to the work was the result of a “mechanical reproduction” or if it represents the author’s own creative conception. However, according to the current guidelines, the USCO will not register works that have been produced by a machine or a mechanical process that operates randomly without any input from a human author.
The possibility of granting copyright protections to AI-generated works has been left open by the office. However, the decision will be based on certain circumstances, such as how the AI tool functions and its role in creating the final work. The office will evaluate each case individually, and if the work lacks human authorship due to being created by a machine, the office will not register it.
In an exciting development last month, the USCO made a decision regarding images produced by Midjourney for a graphic novel. According to the office, these images were not eligible for copyright protection due to the significant “distance” between Kashtanova’s inputs and Midjourney’s output. While the text and layout of Kris Kashtanova’s Zarya of the Dawn were found to be copyrightable, Kashtanova’s lawyers have expressed concern that the office applied the wrong legal standard by focusing solely on the output rather than the input.
The USCO has looked into the intricacies of copyright law and policy issues related to AI per the requests of Congress and the public. The upcoming months of April and May will witness several panel discussions on the subject. The office also aims to gather public comments later in the year on a range of copyright issues surrounding AI usage.
- You can now use Google Translate on the web to convert text from pictures. This development marks a significant advancement in the field of language translation and will prove beneficial for users who frequently access content in different languages they don’t necessarily know.
- Google introduced MusicLM, a model for generating high-fidelity music from text descriptions like “a calming violin melody backed by a distorted guitar riff.” MusicLM models music as a hierarchical sequence that produces consistent 24 kHz music over several minutes.
- Meta colleagues rushed in with the new MakeAVideo approach after generating photos with MakeAScene. It generates full videos from a written description, and users can give a rough description of a scene and make a short movie.
Read more related articles:
In line with the Trust Project guidelines, please note that the information provided on this page is not intended to be and should not be interpreted as legal, tax, investment, financial, or any other form of advice. It is important to only invest what you can afford to lose and to seek independent financial advice if you have any doubts. For further information, we suggest referring to the terms and conditions as well as the help and support pages provided by the issuer or advertiser. MetaversePost is committed to accurate, unbiased reporting, but market conditions are subject to change without notice.