Beyond the Pixel: Understanding the Copyright, Consent and Authenticity Crisis in AI Image Generation

Artificial intelligence (AI) has become so persistent in modern culture that it feels like it has been here forever. Its proliferation, whether through advertising or workplace implementation, is so widespread that it feels inescapable. An AI image generator was one of the first ways for laypeople to dip their toes into generative AI, and it has caused controversy since its inception. From ethical dilemmas to copyright concerns, the pain points seemingly never end. What are all the concerns the public has over AI image generators, and how it is impacting creators?

The Training Data Dilemma of Copyright Infringement on a Massive Scale

Countless people are suing AI companies for having their work stolen to train models. Nobody’s consent was obtained, especially if the content was available online. The leading AI models have stolen millions of books, articles and other works from creators without their consent. The same has happened to images, especially libraries of historical artworks. Copyrighted works of art have also been taken, as evidenced by how easy it is to recreate the world’s most popular and current intellectual properties with a simple spin on them.

In 2023, a case study explored the LAION-5B dataset, which helped train Stability’s Stable Diffusion AI. It has tons of copyrighted images and stock photos, which have sparked countless lawsuits and disputes. Similar actions have occurred with many an AI image generator, such as Midjourney. It is difficult to deny the artists’ allegations — their work was taken and used without compensation or permission.

Who Is the Author of AI-Generated Works?

There are many ethical questions about AI-created images. Who owns the work an AI creates? Is it partially owned by every artist whose work contributed to the piece? Is it the company that created the model? Or is it the person who prompted the AI? Nobody has the answers to these questions yet. However, some terms and conditions from AI companies are attempting to claim ownership over these images, though legislative action could change the trajectory. 

Copyright experts from the Copyright Alliance assert this, however, “If a work is solely generated by an AI and lacks human authorship, there is no copyright protection and therefore no one can own the copyright to the generated work because it is in the public domain.” The U.S. Copyright Office is trying to hammer down the details, but this doesn’t even scratch the surface of this discourse happening across the world.

The End of Artistic Style Due to Mimicry

People have slowly developed an eye for what an AI-generated image looks like, leading to worldwide distrust of what they see. Everything is trying to look like other people’s artwork while mimicking their signature styles.

Fantasy artist Greg Rutowski has developed a style that is extremely identifiable, and with the number of AI prompts stealing the aesthetic, it is making Rutowski’s art harder to spot in the wild. It has led the artist to speak out about how an AI image generator is stealing a skill that took a lifetime to foster, simplifying it into a worse form of AI.

The Illusion of Choice and the Inadequacy of Opt-Out Systems

Countless artists have had their work stolen, so now people are seeking ways to prevent future art from facing the same fate. However, a standardized, widespread opt-out program does not exist. Some AI companies are creating them in response to backlash, but they are not foolproof. Some are outright ineffective, giving artists a false sense of hope that their art remains protected and unused in AI training.

The Economic Impact and Devaluation of Human Artistry and Labor

Many people are going to use free AI image generator programs to create a piece rather than hiring an artist for a commission. The output is good enough, making it a tempting and affordable option. However, many moral and ethical questions arise from these choices. Choosing a program over a human could be seen as devaluing the depth of artistry and labor. This has led to many artists underselling their work just to be competitive with the AI programs.

Surveys show 90% of artists do not trust current laws to protect their employment. Their incomes and stability are at stake because of negligence in keeping up with AI scaling. When the majority of artists believe that scraping the internet for artwork is unethical for AI training, it shows how questionable it can be to replace an artist with a prompt.

The Erosion of Trust in Visual Media Due to Deepfakes

Consent and authenticity are particularly challenging to discern, especially in the era of AI image and video deepfakes. These can look nearly real, potentially damaging an artist’s reputation. This is especially true if something slanderous, defaming or illegal is expressed in these deepfakes. Misinformation and falsehoods proliferate because an AI image-generation tool went through the internet and stole people’s identities. This has even affected the stock market, and it is impossible to predict how it will affect other events, like election cycles.

The Transparency Issue of the Black Box Problem in AI Models

Unless a company is deliberately incorporating explainable AI, the image-generation program has a black-box problem. Nobody can tell where each element of an AI-generated image comes from. The picture does not clarify who or where the picture came from, and in what way it influenced the output. This makes it impossible to cite sources or identify which facets of it were pulled from reality. There is no way to track provenance, making the issue all the more complicated. 

Toward a More Ethical Digital Canvas

The sheer variety and moral depth of all these concerns will continue to plague viewers and makers of art for years to come. However, legislative action can assist to regulate AI. Proper training needs to be standardized. Similarly, people need options to opt out of these systems at any stage of the process. 

Additionally, everyone needs more education on the ethical use of AI image generators. They are causing widespread confusion about the value of art and the people who have created works for thousands of years. To preserve artistic merit, the miracle of human ingenuity and set a sustainable precedent for use, humanity must design considerate systems for AI training and use.