Familiarize yourself with copyright laws and intellectual property protections in your region.

Copyright protects original works of authorship, granting the creator exclusive rights to reproduce, distribute, and adapt their creations. Traditional copyright laws attribute authorship to human creators, which presents new challenges. Some jurisdictions, like the United States currently excludes no-human authorship from copyright protections. Works that are created with significant human input, using AI as a tool, are generally eligible for copyright protection. The level to which human involvement is required to claim authorship remains a consideration. AI systems often rely on large datasets for training which may include copyrighted material. Legal disputes have arisen over whether such data constitutes as fair use or infringement.

Intellectual property (IP) laws also extend to AI creations. AI can generate branding elements such as logos and slogans which can be trademarked if they meet a distinctiveness criteria. 

Copyright laws for AI: https://www.copyright.gov/ai/

Intellectual property protection resources: https://www.uspto.gov/trademarks/basics/trademark-patent-copyright

As AI systems grow to interact with personal data and creative works, the ability to opt-out of certain uses of AI has emerged as a critical issue. Opt-out options provide individuals, organizations, and creators a pathway to gain control over how their data and intellectual property are utilized by AI systems.

Monitor platforms offering opt-out mechanisms for including your art in AI training datasets (ex:  Stable Diffusion's opt-out protocols).

https://www.technologyreview.com/2022/12/16/1065247/artists-can-now-opt-out-of-the-next-version-of-stable-diffusion/

Other platforms such as social media (ex: X, Instagram, Reddit, Tiktok) have options within their settings that allow opting out of being used in data sets.  

Linked below is information on how to perform opting out for each.

https://www.techtarget.com/whatis/feature/How-to-opt-out-of-AI-training-across-social-media-platforms

Use metadata or digital watermarks in your artworks to assert rights and discourage unauthorized use.

Additional resources on how to use metadata, watermarks, and other available tools such as disabling downloads linked below

https://businessofillustration.com/metadata-watermarks-image-protection/

Licensing and attribution are critical concepts in the context of AI, shaping how systems use data, generate content, and respect intellectual property rights. These frameworks ensure that creators maintain control over their works while enabling AI developers to access needed resources. Licensing refers to the agreements governing the use of content, data, or technology in AI systems. There are various types of licenses from open source, which are often used in public datasets to train AI to custom licenses that are tailored agreements for AI applications

For more information on types of licenses and their usage permits and restrictions please visit:

https://viso.ai/deep-learning/ai-licenses

Attribution involves giving proper credit to creditors whose works contribute to AI-generated systems and ensures acknowledgment of original contributions and adherence to licensing requirements.

Further information on attributions can be found here.

https://fullcircleinsights.com/blog/ai-in-attribution/

Clearly define the terms of use for your artwork through licensing agreements

Require attribution if your work is used for derivative creations or included in training datasets. 

Monitoring outputs of AI systems is essential to ensure that AI-generated content adheres to legal, ethical, and creative regulatory standards. As AI becomes more integrated in the production of art, music, literature, and other forms of creative expression, oversight is necessary to address issues related to originality, cultural sensitivity, and intellectual property.

Use reverse image search (Google or other) or AI detection tools to track unauthorized use or replication of your works in AI-generated art periodically. 

There are various software available for AI detection and a compiled list can be found here: 

 https://images.google.com/

 https://guides.library.ttu.edu/artificialintelligencetools/detection

Frequently reference sites that track if AI has been using your material in training sets with sites such as:

https://haveibeentrained.com/

Transparency in AI is crucial for building trust, ensuring accountability, and fostering ethical development and use. Advocacy and transparency involves promoting practices that make AI systems, their data, and decision making processes understandable and accessible to all.

Support initiatives and legislation that demand transparency in how AI models source and use training data.

Further ethics, advocacy, and critical evaluation resources can be found:

https://guides.library.berklee.edu/artificial-intelligence/ethics-advocacy

Engage with artist communities advocating for ethical AI practices. 

Additional information on transparency can be found:

 https://www.zendesk.com/blog/ai-transparency/