Copyright challenges in the age of AI: Who owns AI-generated content?

The FOSSA Podcast covers engineering-product team collaboration (and friction), product management tools, when to hire your first PM, and more. However, after being denied protection each time, Thaler sued the Copyright Office in June 2022. And although it does have its limitations, generative AI can be certainly be leveraged to strengthen our creative proposals and productivity, leading to faster turnaround. Any personally identifiable information you share with Northern Light will be used only for purposes of processing your transaction. We will not give, sell, or rent your name, e-mail address, credit card numbers, mailing address, purchasing history or any other personally identifiable fact we learn about you to a third party.

Those uses do nothing to further learning, and actually pollute public discourse rather than enhance it. We believe that while inputs as training data is largely justifiable as fair use, it is entirely possible that certain outputs may cross the line into infringement. In some cases, a generative AI tool can fall into the trap of memorizing inputs such that it produces outputs that are essentially identical to a given input. “Denying copyright to AI-created works would thus go against the well-worn principle that ‘[c]opyright protection extends to all ‘original works of authorship fixed in any tangible medium’ of expression,” Thaler said. Concerns about the impact of new technology on human creators and calls to impose IP-based restrictions on emerging technology are not new.

While generating the right prompt for the piece demanded hundreds of different prompts from Allen – with the process as a whole taking more than 80 hours – the AI image was considered by many not worthy of competing with human creation. The findings corroborate existing worries of copyright infringements in the world of generative AI, and the researcher warned that… It’s “virtually impossible” to verify that an image created with Stable Diffusion is original and “not stolen from the training set”. The ruling marks the most recent volley in a series of disputes between Dr. Stephen Thaler, a computer scientist, and the world’s prominent intellectual property regimes.

Generative AI and Copyright

Some have argued that the use of training data in this context is not a fair use, and is not truly a “non-expressive use” because generative AI tools produce new works based on data from originals and because these new works could in theory serve as market competitors for works they are trained on. While it is a fair point that generative AI is markedly different from those earlier technologies because of these outputs, the point also conflates the question of inputs and outputs. In our view, e using copyrighted works as inputs to develop a generative Yakov Livshits AI tool is generally not infringement, but this does not mean that the tool’s outputs can’t infringe existing copyrights. Artwork created by artificial intelligence isn’t eligible for copyright protection because it lacks human authorship, a Washington, D.C., federal judge decided Friday. The Copyright Office will not register works whose traditional elements of authorship are produced solely by a machine, such as when an AI technology receives a prompt from a human and generates complex written, visual or musical works in response.

The European Union, which has a much more preemptive approach to legislation than the U.S., is in the process of drafting a sweeping AI Act that will address a lot of the concerns with generative AI. And it already has a legislative framework for text and data mining that allows only nonprofits and universities to freely scrape the internet without consent — not companies. Like most other machine learning models, they work by identifying and replicating patterns in data. So, in order to generate an output like a written sentence or picture, it must first learn from the real work of actual humans.

Does generative AI violate copyright laws?

Similarly, in the same month, the comedian and writer Sarah Silverman and authors Christopher Golden and Richard Kadrey claimed that both OpenAI and Meta’s models were trained using their work without permission. The authors have filed a lawsuit against OpenAI and Meta, claiming that the companies violated copyright law by using their material without obtaining permission to train the AI models. The NOI seeks factual information and views on a number of copyright issues raised by recent advances in generative AI.

• Developing these audit trails would assure companies are prepared if (or, more likely, when) customers start including demands for them in contracts as a form of insurance that the vendor’s works aren’t willfully, or unintentionally, derivative without authorization.
• For example, while each Output Work may be unique, the generation process can result in Output Works that are substantially similar to Input Works.
• Given the international nature of the Internet, there is some risk that documentation requirements will become de facto global requirements.
• While the technology is being hailed within the marketing industry for its ability to supercharge and supplement human creativity, it’s also presenting some thorny legal questions.
• There is some nuance in this, of course, as the specificity of prompts varies substantially.

Allen filed an application for copyright registration but did not disclose Midjourney’s role. The Copyright Office refused to register the work because Allen declined the examiner’s request to disclaim portions of the artwork generated by AI. Various jurisdictions around the world are beginning to address the copyright issues relating to AI. Japan and Singapore have enacted specific AI exceptions that do not require compensation.

Yakov Livshits
Founder of the DevEducation project
A prolific businessman and investor, and the founder of several large companies in Israel, the USA and the UAE, Yakov’s corporation comprises over 2,000 employees all over the world. He graduated from the University of Oxford in the UK and Technion in Israel, before moving on to study complex systems science at NECSI in the USA. Yakov has a Masters in Software Development.

Some leading firms have created generative AI check lists for contract modifications for their clients that assess each clause for AI implications in order to reduce unintended risks of use. Organizations that use generative AI, or work with vendors that do, should keep their legal counsel abreast of the scope and nature of that use as the law will continue to evolve rapidly. The USCO’s decision has major implications — and creates potentially significant challenges — for engineering teams. It would require developers to distinguish between code they wrote with and without generative AI, which is often impractical. The idea of GANs is that we have two neural networks, a generator and a discriminator, which learn from each other to generate realistic samples from data. Regardless of who is right, it is very odd for fanfiction writers, who rely on fair use to justify their use of characters created by others, to turn around and claim that others are not to make fair uses of their creations.

💡 A registry for AI-generated content and authors gains traction as a potential solution. Generative AI has revolutionized content creation, but attributing contributions to individual authors becomes difficult due to the amalgamation of vast datasets from diverse sources. The guidance also outlines the responsibilities of copyright applicants to disclose the use of AI-generated content in their works, providing instructions on submitting applications for works containing AI-generated material and advising on correcting a previously submitted or pending application. The Copyright Office emphasizes the need for accurate information regarding AI-generated content in submitted works and the potential consequences of failing to provide such information.

AI models work by deriving abstract patterns and relationships from billions of pieces of training data, and using those abstract correlations to create wholly new content. They are not designed to reproduce protected material from the data on which they are trained—and on the rare occasions that they do, copyright law provides the tools necessary for courts to enforce rightsholders’ legitimate protections. The ruling has implications for generative AI and users of AI tools like ChatGPT, Midjourney, and DALL-E. Within that context, we see generative AI as raising three separate and distinct legal questions.

Much of this coverage contains serious inaccuracies about AI technology and copyright law. The issues surrounding AI and copyright law can be complex, therefore we’ve collected a number of the more prevalent misconceptions in recent media and explained why they are false to aid in the conversation around this technology. In addition, fair use of copyrighted works as training data for generative AI has several practical implications for the public utility of these tools.

Can Generative AI Already Do Basic Legal Tasks as Well as Lawyers?

These licenses include terms that dictate the public’s ability to use Wikipedia text, including “share alike” provisions that require works that alter, transform or build upon Wikipedia works be distributed under the same, similar or compatible license schemes. Under limited fair use maximalism, Output Works generated from GAIs trained on Wikipedia articles would be subject to the same “share alike” provisions. Sitting between the two extremes is what we call conditional fair use maximalism – an approach that evaluates an Output Work on a case-by-case basis to determine whether the fair use defense should apply.

Many lawsuits have already been filed against AI image generators that contain copyrighted images in their training data. Considering one of the biggest challenges to copyrighting AI-generated content is the possibility of copyrighted material being used to train the AI system, labeling it could be a step in the right direction that will potentially lead to more refined copyright laws in relation to AI-generated content. In the long run, AI developers will need to take initiative about the ways they source their data — and investors need to know the origin of the data. Stable Diffusion, Midjourney and others have created their models based on the LAION-5B dataset, which contains almost six billion tagged images compiled from scraping the web indiscriminately, and is known to include substantial number of copyrighted creations.

The Office earlier this year held a series of “listening sessions” with stakeholders, including representatives of Microsoft (a major backer of OpenAI), VC firm Andreessen Horowitz and The Authors Guild. The Office is looking at possible regulatory action or new federal rules due to “widespread public debate about what these systems may mean for the future of creative industries.” On the left, Dall‧E was asked to generate an image of “an astronaut riding a horse in a photorealistic style.” On the right, Dall‧E was also asked to generate an image of “an astronaut riding a horse,” but this time it was asked to do so “in the style of Andy Warhol.” Senate, expressing concern about calls for new copyright legislation that would jeopardize the benefits of AI and upend the core governing principles of our nation’s intellectual property regime. There are several actions that could trigger this block including submitting a certain word or phrase, a SQL command or malformed data.