Pa ge 1 Pa ge 46 American Journal of Society and Law ( AJSL) Impact of Generative Artificial Intelligence on The Global Entertainment Industry & Its Legal Dynamics D. O. Ben-Daniel1* Volume 4 Issue 1, Year 2025 ISSN: 2835-3277 (Online) DOI: https://doi.org/10.54536/ajsl.v4i1.4566 https://journals.e-palli.com/home/index.php/ajsl Article Information ABSTRACT Received: February 20, 2025 Accepted: March 21, 2025 Published: April 28, 2025 The emergence of Artificial Intelligence (AI) has given rise to unprecedented and momentous consequences across humanity. There is no doubt that AI, most especially, Generative (Gen) AI portents massive consequences within the entertainment industry. This article attempts to contribute to the overall discuss of gen AI within the entertainment industry, and covers the proper understanding of gen AI and how AI models are trained; the entertainment industry; the impact of gen AI on the entertainment industry globally; the advantages and disadvantages of deploying gen AI within the entertainment industry; and the legal tools available to manage the existential threat of gen AI to the entertainment industry. Therefore, the article explains and discusses AI broadly, gen AI specifically, the training process, the current developments within the entertainment industry regarding the adoption of AI, and the legal consequences of gen AI, such as, copyright infringements. This article concludes by emphasizing the quantum advantage of deploying gen AI as an effective tool within the entertainment industry, which shifts the focus from the threats gen AI portends for the industry at large, to its huge benefits if properly harnessed. Keywords Copyright, Creative Industry, Entertainment, GenAI, IP 1 7220 McCallum Blvd, Dallas 75252, Texas, USA * Corresponding author’s e-mail: gbendaniel@gmail.com INTRODUCTION The past two decades have witnessed exponential growth in machine learning and robotic science, known as artificial intelligence (AI). The growth and adoption of these technologies across multiple industries have been unprecedented, to the extent that there has been a growing fear that these technologies could eventually replace humans or even render humans “useless”, as if humans did not develop these technologies. This global phenomenon is inferencing the possibility that AIs could finalize the much touted replacement theory, a similar sentiment expressed by a leading scientist, Joseph Weizenbaum, that the study of AI is obscene, anti- human and immoral. This fear of human replacement has always been expressed, each time humanity witnesses an evolution or revolution in technology, right from the age of industrial revolution as we know it today. Considering the pace of development and adoption of AI technologies across industries, such replacement perception is not unfounded. Most especially, with the release of AI engines like ChatGPT, Gemini, Adobe Photoshop (powered AI), and the most recently released Chinese AI, DeepSeek, that crashed valuations of AI companies across global stock markets. Furthermore, the simultaneous crashing of the computing power requirements of these Large Language Models (LLMs) is leading to more computing power becoming available at extremely affordable costs, increasing the rate of adoption across industries. Adopting the AI phenomenon within the entertainment industry has also produced a monumental impact within the industry, such that the same replacement theory is being touted, leading to threats of strikes, law suits, copyright infringements, etc. If not properly managed and streamlined, gen AI could be more damaging than constructive and developmental to the global entertainment industry. Therefore, the objective of this journal write up is not to pitch “Humans vs AIs” narrative in an adversarial way, but to demystify this global phenomenon, analyse its growth and impact within the entertainment industry by showcasing the gen AI benefits, and the potential threats to the industry if gen AI is not properly harnessed. Furthermore, unveil the opportunities available within the copyright and entertainment laws when gen AI is properly and strategically utilized and deployed. Research Problem To what extent has the emergence of gen AI impacted the global entertainment industry, and to what extent does strategic integration of gen AI tools into the entertainment industry’s value chain directly correlate with increase in operational efficiency, creativity and overall value of the industry. Research Questions • How does the adaptation of gen AI automation tools impact the time and cost associated with productions within the industry? • What is the impact of gen AI on resource allocation and waste reduction within the production process? • How does the adoption of gen AI affect the speed and accuracy of script selection process and other processes in production? • What are the quantifiable improvements in productivity and output resulting from the strategic integration of gen AI automation in production process? • How does the adoption of gen AI impact creativity and the overall value of the entertainment industry Pa ge 47 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 Research Objectives • To assess the overall impact of gen AI on the global entertainment industry, its operational efficiency, creativity and overall profitability. • To analyse the impact of this phenomenon on the human creatives (workforce) within the entertainment industry, and the changing dynamics because of adoption of gen AI within the industry. • To evaluate the ethical and societal impact of gen AI on the industry, the legal instruments available to limit the “damaging” effect of the phenomenon within the entertainment industry. • To evaluate the cost-benefit analysis of the adoption of gen AI within the global entertainment industry. LITERATURE REVIEW This article clearly articulates the established concepts of AI and description of the entertainment industry. The Oxford Languages defines AI as the theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages. According to Stryker & Kavlakoglu (2024), AI is a technology that leverages on smart computers and machines to simulate human learning, comprehension, problem solving, decision making, creativity and autonomy. Turing (1950), an English mathematician and a leading AI researcher, first conceptualized the “thinking computers or machines,” acknowledging that only a special kind of machine, known as “digital computers,” could do that fit. McCathy (2007), a computer scientist at Stanford University and one of the founding fathers of AI, defines AI as the science of making intelligent machines and software programs, adapting them to human intelligence, without necessarily confining it to methods of biological observations. He holds that machine intelligence, though adaptable to human intelligence, requires some mechanisms of intelligence that enables it deliver impressive performances on the tasks its built for. Colorado State University (CSU) Global (2021) sates that AI technology enables computers and machines to mimic human intelligence through an interactive processing and algorithm training. CSU further explains that the combination of interactive processing and algorithm training enables problem- solving and providing strategic directions, which is a process of continuously and intelligently relying on its interactive algorithms through the combination of large data sets that are analyzed and then learning from their patterns and features. Eventually, the machine develops high capabilities and learnings from these data patterns, which enables it to intelligently answer or solve problems, and possibly project (predict) into the feature based on these data patterns. To fully comprehend the concept of AI and its applications, the component parts must be properly understood. According to CSU, these component parts sum up the discipline of AI, and they make up the sub- domains of the main AI domain. Some of the sub- domains of AI are machine learning (ML), which is component of AI that enables computer systems and programs to learn real time, and advance outcomes based on its experience, rather than how it’s programmed. This is where AI discovers data patterns and logic that enable it deliver on its pre-set outcomes or deliverables. The other one is deep learning, which is the type of ML enables AI improve by processing data, because it utilizes artificial neural networks that mimics the human brain’s biological neural networks when processing information. Others are the neural network itself that functions like a human brain, but with capacity to work with large dataset. These make up the entire global AI architecture that drives this cutting edge technology across multiple sectors like healthcare, legal services, financial services, manufacturing, creative (including entertainment), automobile, etc. Stryker & Kavlakoglu (2024) affirm that AI has evolved over the decades from its inception in the 1950s from machine learning (where machines learns from historical data), to deep learning (where machine learning models try to mimic humans’ brain function), and now, generative (gen) AI (where deep learning models can create original content, such as text, videos, images, etc.). Gen AI is that AI used to create new content such as text, videos, images, music, computer codes and audio. Gen AI is built on two components of AI, namely, machine learning and deep learning. Stryker & Kavlakoglu (2024) emphasise that understanding these two components is integral to comprehending gen AI. Gen AI deploys these two component to identify and learn patterns and relationships in a dataset provided, and then utilizes these learnings to generate new content. The newly created content could be images, texts (inform of essays or chat responses), videos, audios, and computer codes. The examples of gen AI are Chat GPT, DALL-E 2, Gemini, DeepSeek, etc. Therefore, gen AI is trained on provided dataset, through the process of machine learning and deep learning, its able to understand the dataset patterns and relationships, and be able to develop its own ability to generate new content. Gen AIs like Gemini, Chat GPT, DeepSeek, etc., have been trained on large volume of dataset, which explains their ability to generate new content as requested by users. Gen AI has various learning processes, such as supervised, which is a technique of machine/deep learning that utilizes labeled dataset in its learning process of identifying data patterns and their relationships. Toloka (2023) explains that labeled dataset contains meaningful tags (information) that requires extra process of labeling (such as specifying what object is in the image dataset, be it car or bird; or even the words uttered in the recording, etc.) This makes the dataset more meaningful to the machine. Labeling of dataset is a fundamental step towards building a high-performing and accurate gen AI algorithm, which is the goal of its learning process-to develop an AI model that delivers correct output based on real-world data. The objective of this is to ensure an effective and “supervised” training process that guarantees a more accurate result. Pa ge 48 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 Explaining further that these labels play a vital function in enabling the gen AI model deliver the best outcome and predictions, for example, predicting travel time, based on the time of the days and the weather conditions. The other type of learning is unsupervised learning, and IBM (2021) states that this relies on machine- learning (ML) algorithms to identify hidden data pattern and groupings without human intervention. The ML algorithms analyzes and clusters unlabeled datasets, and identifies the similarities and differences of the data (information) contained in the dataset, and this type of gen AI is ideal for data analysis, customer segmentation, image recognition (object recognition), and cross selling strategies. According to Bergmann (2023), another type is semi-supervised learning, combines both supervised and un-supervised learning, thus giving a bit of both worlds. It utilizes both labeled and unlabeled datasets for its training for both classification and regression tasks. This solves the problems of not having enough labeled dataset to utilize for learning, and when it’s too expensive to label enough dataset. Another type of learning is self-supervised learning that uses unsupervised learning to train models that require supervised learning, mostly because of the difficulty and time required to label the dataset. Thus, self-supervised learning is more time-effective and cost- effective because the need to manually label the dataset is replaced with an AI model generating implicit labels from unstructured data. Another type of learning, as analyzed by IBM, is reinforcement learning, which is similar to supervised learning but not trained using a sample dataset. It is rather based on trial and error, and the successful sequence of outcomes is reinforced to develop the best recommendation or policy for a given problem. Transfer learning, which Murel & Kavlakoglu (2024) explain to be the type of learning that utilizes pre-trained AI models and dataset to optimize the performance and generalizability for a related task or dataset. The product or output of these trainings is what is know as large language model (LLM) that now functions as a base model for any gen AI. These base models can further be trained for industry vertical or specific performance. This means, some LLMs are ready for use after training, such as Chat GPT, Gemini, DeepSeek, etc., whilst, some are just base model (framework) to be further trained vertically to perform (generate new content) industry specific tasks, some of which are GPT-3, LLaMA, Gemini, DeepSeek, etc. These base models have been pre-trained on vast amount of data for generalizability and related tasks, and they are just retrained to further perform narrower related and specific tasks. For example, retraining a GPT- 3 LLM model to perform legal specific functions like legal writing, case summarization, etc., or even training the same model to function within a hospital setting, such as patient data analysis for tracking and diagnosis of chronic diseases, diseases outbreak predictability, etc. On the other hand, the entertainment industry is an economic sector made up of individuals and businesses that engage in creativity, such as designs, music, motion pictures, gaming, publishing and literature, visual arts, performing arts, fashion, TV and radio, and advertising. The entertainment industry is made up of individuals (humans) with creative skills and talent that are creating a thriving economy through their work of designs, music, motion pictures, writing, performing arts, etc. According to PwC (2024), the Media and Entertainment is expected to hit a global valuation of US$3.4 trillion in 2028, and the National Assembly of State Arts Agencies (NASAA) confirmed that the U.S. entertainment industry alone was valued US$1.1 trillion as at the end of 2022. The entertainment industry is one of the fastest growing industry, globally, most especially with its massive potential for wealth and job creation. In most developing/emerging countries like India, Nigeria, Ghana, South Africa, Brazil, Turkey, Mexico, Peru, Cuba, etc., the entertainment industry is one of the greatest wealth creators and employers. Over the decades, the entertainment (creative) industry has always depended on intellectual property (IP) law to protect its creation, and maximize both its value and return on investments. The works of the entertainment industry, such as, designs, music, motion pictures, gaming, publishing and literature, visual arts, performing arts, fashion, TV and radio, and advertising are called “original works” or “creative works” in the copyright law world, and they are copyright protected (patent protected in certain jurisdictions). What makes the entertainment industry function effectively is its ‘regulation’ by the intellectual property legal instruments, such as copyright, patent, designs, trade secrets; and other legal instruments such as contracts, torts, and personal rights. MATERIALS AND DISCUSSIONS Research Design This research design adopts qualitative research method, which involves adopting emerging case studies within the entertainment industry; entertainment industry insider’s perspective (ethnography), which involves understanding inside dynamics within that industry; analysis of lived experiences of individual players within the industry; and the established legal theories regulating the industry. Data Sources The research work was sourced from the U.S. constitution, statutory instruments, regulations, published journals of other authors, researched works and news reports of credible media outlets, and other verified authorities. Consistently ensuring legal and other verified authorities constitute the sources for this research ensures the authenticity and accuracy of the data analysed to arrive at the conclusions of the research work. Analysis This research work involves the textual analysis of constitutional, statutory and regulatory frameworks that define and establishes the legal position in copyright law, Pa ge 49 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 which influences the legal framework for the entertainment industry. These textual analysis is then applied to the current trends and dynamics within the entertainment industry, as its been revolutionized by gen AI. This then informs the conclusions and recommendation discussed in the article. RESULTS AND DISCUSSIONS The past 2-5 years have seen significant adoption of gen AI within the entertainment industry, leading to definitive impact on the industry. The architecture of gen AI is generally to mimicking humans, which ensures the execution of the same tasks that humans do, but at a larger and more efficient scale. Several gen AI applications have been developed for the entertainment vertical sector, which has had significant impact, largely leading to broken trust amongst the industry stakeholders. In consistent with gen AI’s training processes described above, Kinder (2024) explicated that a gen AI model can be trained with vast amount of existing movie scripts, and based on this training, can generate new scripts based on ideas, dialogue, types of characters, and even plot developments that appear to work well together. This will have massive impact on script development within the entertainment industry, both positively and negatively. The benefits and costs of gen AI to the entertainment industry is discussed later in this article. The summary of this point is that a gen AI model, pretrained on vast amount of existing scripts, can subsequently create new scripts, not requiring human script developers any longer. In many instances now, gen AI creates the idea and the scripts, and the writers are simply hired to polish and re-write the scripts, which is much better, as articulated further in this article. This development led to the breakdown of trade agreement between Writers Guild of America (WGA) and The Alliance of Motion Picture and Television Producers (AMPTP), the alliance of over 350 American television and film production companies in May 2023, leading to a 148-day strike against the use of AI in script development. Sakoui confirms that the Screen Actors Guild-American Federation of Television and Radio Artists (SAG-AFTRA), and Teamsters and International Alliance of Theatrical Stage Employees (IATSE) also joined the strike by WGA, against AMPTP because of the same perceived threat from gen AI. Another big controversy generated by gen AI, as reported by Korn, (2023), is in the area of image generation that involved Getty Images suing Stability AI for copyright violations, regarding its arts tool, Stable Diffusion. Getty accused Stability of illegally training its AI model utilizing Getty’s images without obtaining license to do that. According to by Lan (2023), the legal challenges emanating from “AI painting” alone grew by 560% between July 2022 and January 2023. This percentage growth in the volume of related litigations on copyright infringements within the entertainment/creative industry depicts the dimensions gen AI is taking the industry to in the coming years. Another dimension is the application of gen AI to voice generation, and the one main example, as reported by Ingham (2023), is an AI song featuring fake Drake and the Weeknd vocals that showed up across multiple streaming platforms of YouTube, Spotify, TikTok, Tidal and Apple Music. This triggered prompt response from Universal Music Group (UMG) on April 17, 2023, requesting the streaming platforms to pull down immediately the AI song from their respective platforms. This is the power of gen AI, trained with the vocals of known personalities like celebrities, politicians, etc., and then deploying it to create fake audio vocals over videos, be it music, news, or motion pictures. This trend is multiplying exponentially, with enormous impact in deceiving global audiences and the marketplace, denying artists and other right owners royalties and other revenues due to them. From the forgoing, the impact of gen AI on the entertainment industry is extremely significant, and has both advantages and disadvantages. The advantages of adopting gen AI within the industry include the following: Pace of Development The overall pace of developing a creative work is enhanced with gen AI. In the traditional mode, developing a movie production could take years, because of the intense work required to develop every part of the production. Script development is part of the pre-production phase, and it’s a fundamental stage of the production, and could take months/years. Traditionally, the pre-production stage of script acquisition or development could require several options, namely, securing a copyright protected pre-developed script, or secure a book to develop into a script, or getting skilled scriptwriters to develop fresh scripts. Any of the options selected, though tedious and fundamental to the production process, can still be optimized and enhanced by gen AI. Without an effective script development process, you might not have an authentic and quality production, worthy of generating the revenue numbers for all the stakeholders in the project. The other pre-production activities are artists/ cast selections, photoshoot, crew selections, etc., and all of these can be fast tracked and optimized deploying gen AI. Cost and Timing of Production The first benefit of gen AI to the entertainment industry is the potential cut in cost of production, and the time it currently takes to complete a motion pictures production of any kind, which includes the pre-production, the actual production and the post-production. This applies across the entertainment industry, namely, music (both audios and videos), performing arts, gaming, etc. Gen AI’s ability to save cost and timing could actually help remove barrier to entry within the entertainment industry. According to Nashville Film Institute, major global studios spend between $70Million and $150Million to make a feature film, covering the cost of crew/cast, scriptwriters, producers and directors, logistics, and post- production. However, Movie Budgets affirms that some Pa ge 50 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 blockbuster movies are produced as high as $600Million. No doubt, the African country with the biggest creative industry is Nigeria, and from experience, the average cost of producing a motion picture in Nigeria range from as low as $2,000 to $300,000, depending on casts/crew, production value, target audience and platform, logistics, etc. However, with gen AI, according to Marcolivio (2024), the average cost of producing a digital movie could go as little as $829 an hour, eliminating cost of logistics, insurance, lowering or eliminating cost of crews/casts, etc. This completely removes cost-induced barrier to entry for movie makers, and massive reduction of the time to produce to just about a few days for an hour production. Improved Efficiencies & Elimination of Wastes Embedded in the cost and time savings above is gen AI’s ability to improve efficiencies in the production of a motion picture, music, or creative arts, and help eliminate waste. Marcolivio (2024) explains that AI could eliminate or reduce the time spent on pre-production activities such as scenes set ups, shoot days, night scenes dynamics, delays from bad weather, and shooting of scenes with animals. Gen AI guarantees operational efficiencies and waste elimination, because every part of the creative process is optimised and automated, which ensures quick turnaround and effectiveness. An example here is Jumpcut’s AI tool used to read through 50,000+ submitted scripts, and generates two-page report for producers and directors, breaking it down into genre, subgenre, characters and similar titles. This simply does not replace human creativity; it just enhances script selection process, saves time and improves quality. Emerging New Business Models Gen AI within the entertainment industry is driving new business models within the industry. As the AI phenomenon scales up, new AI character (actors) ecosystem is emerging, outside of the traditional entertainment’s characters (actors) ecosystem. This involves creating digital actors whose features are programmed to meet individual preferences, almost the same patterns or similarities with fictional characters. The digital actors take several formats such as super hero, a training coach, a newscaster, an avatar, music super star, TV show presenter, etc. To exploit this emerging digital character creation, new business models are emerging within the AI powered entertainment industry. Marcolivio (2024) gave a few recent examples; Meta paid Snoop Dogg $5million to create AI personae, Lucasfilm signed a $22 Million deal with James Earl Jones to recreate Darth Vader voice with AI. Others are Cadbury powered AI that features Indian super star, Shah Rukh Khan, etc. In addition to the digital assets’ creation is the potential for marketplaces for AI characters (actors) to emerge in the coming months that would be sources of characters (actors) for the entertainment industry. On the flip side, the disadvantages of adopting gen AI within the entertainment industry are also significant, compared with the benefits: Economic Impact on the Industry According to the PwC’s report, the global valuation of the entertainment industry is expected to hit US$3.4 trillion. This valuation covers all the value chain within the entertainment industry, such as script development, other production preps, photography, video editing, main production, postproduction, and marketing & distribution. This valuation spreads across the value chain, though not evenly distributed, but every part of the value chain retains substantial valuation. However, Gen AI potentially could distort that spread, thus, creating massive disruptions within the industry, and could potentially wipe out value across the different subsectors of the entertainment industry. Gen AI, if not well regulated and managed, could have this negative economic impact within the entertainment industry. However, if well leveraged, could leapfrog the entire industry’s ecosystem, triggering the next wave of development within the industry, which has a multiplier effect on the growth and valuations of the industry. This requires consorted and strategic collaborations covering training and development, and adoption of AI as an effective tool for the creative professionals within the entertainment industry. Impact on Copyright Violation and Infringement Another con of gen AI within the entertainment industry is the massive copyright infringements that have taken place during the training of several AI models. Most of the AI models have been trained with copyright protected materials such as images, scripts, books, audio, video, etc. The fall out of these copyright infringements has led to unprecedented legal battles across the globe. Many more legal battles are still brewing up. The reality is, as we have discussed above, gen AI depends on being pre- trained by huge amount of dataset to be able to function, run effectively, and generating new output based on the historical data it had been trained with. Therefore, most gen AI models currently being deployed within the entertainment industry have been pre-trained with huge amount of dataset from within and outside the industry, some or most of which are copyright protected, thus, infringing on the right holders substantially. Zero Copyright Ownership and Protection Several sections of law are present in the entertainment industry, namely copyright, trademark, “rights to publicity”, contract, licensing, employment, immigration, bankruptcy, defamation, privacy, and torts. However, fundamental to the industry is copyright law, because this is what protects all the creatives of the industry, and provides the foundation for the effective monetization of all created original works of the industry. The U.S. Constitution states that “the Congress shall have the Power… To promote the Progress of Science….by securing for limited Times Pa ge 51 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 to Authors….the exclusive Rights to their…Writings. Then the U.S. Congress responded with the Copyright Act of 1976, which states “original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. Works of authorship include the following categories: (1) literary works; (2) musical works, including any accompanying words; (3) dramatic works, including any accompanying music; (4) pantomimes and choreographic works; (5) pictorial, graphic, and sculptural works; (6) motion pictures and other audiovisual works; (7) sound recordings; and (8) architectural works. (b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work” Ss(b) simply differentiates copyright protection, which deals with works of expressions, from that of patent, which protects ideas, procedure, process, etc. The focus here is the emphasis on “Author” and “works of authorship”. This clearly shows that authors under copyright law are humans, and not a machine or software (AI algorithm) that generates the “work.” Furthermore, the U.S. Copyright Office defines “an original work of authorship is a work that is independently created by a human author and possesses at least some minimal degree of creativity.” The AI generated output is not classified as an “original works of authorship” that is worthy of copyright protection or any form of protection under entertainment law. The main issue here is that AI generated creatives are actually derivative works of several other original works that belong to other authors, which were used to train the AI model based on a copyright doctrine of fair use. In Burrow- Giles Lithographic Co., the court also agreed with the doctrine that only humans qualify as “author” and the ONLY one who can get copyright protection under “original works of authorship.” In the Federal Register, the U.S. Copyright office issued a guideline in regarding AI generated works, stating that when AI “determines the expressive elements of its output, the generated material is not the product of human authorship.” In a recent decision, Stephen Thaler, the court held that “Human authorship is a bedrock requirement of copyright.” This further affirms the legal position that AI generated works, including within the entertainment industry lacks any form of IP (copyright) protection. This similar position is held in South Korea, the European Union (Matt-2023), South Africa and most other African states. However, AI generated materials may be protected in the United Kingdom and China (Wang-2024). The implication of zero copyright ownership and protection means No Extraction for Works of Original Authorship. Without Extraction for Works of Original Authorship the entire entertainment industry structure and value chain is of minimal value. According to the U.S. Copyright Act (§§ 106), an author has the exclusive derivative rights over his/her Works of Original Authorship, which are right to reproduce, distribute, perform, display, license, and to prepare derivative works based on the copyrighted work. Therefore, lack of protection for AI generated materials, means no further extraction of rights (derivative rights) on the AI created works, which limits the ability of the AI created works to maximize benefits to the owners through maximum derivative rights. This is counterproductive for the entertainment industry, which depends on those derivative rights to further maximize value for the industry. Therefore, there needs to be a more structured and planned deployment of gen AI technologies within the industry to avoid gen AI deployments ending up being self-destructive for everyone within the industry. Limited Originality & Uniqueness As described above gen AI depends on historical data, and then generates new content based on the training it has had with the vast amount of historical data. The challenge here is the generic nature of the content generated by gen AI, because of gen AI’s inability to generate unique, original and authentic content for a particular user. Some may argue that the limitation is as a result of lack of proper prompt engineering skills by users, to be able to maximize the uniqueness of a gen AI model. However, as long as gen AI depends on its training with historical data, to generates new output based on these historical data, its output is still largely dependent on its dataset training, which will still have generic output applicable to several users simultaneously. Affirming this argument, Doshi & Hauser (2024) explained that AI enabled stories are more similar to each other, compared to stories generated by humans alone, and this limits creativity, freshness and originality. However, despite the limitation of gen AI, it still helps writers that lack writing skills to write more creatively and more enjoyable. If writers can leverage on stories generated by AI as the starting block, and further develop the stories by infusing their own expressions, uniqueness, and authenticity. Authentic storylines have the elements of uniqueness, novelty and relatedness, which resonates more with the audience. No doubt, gen AI could enhance creativity of a writer, and can also inhibit creativity and originality in another. CONCLUSION From the fore going, the two fundamental legal dynamics highlighted here are, first, the challenges facing copyright holders globally, which is the continuous copyright infringement of their works. There has been increased vigilance by right holders, constantly “scanning” and monitoring new AI models for possible copyright infringements. The Recording Industry Association of America (RIAA) has reported en masse violations of their members’ right through the training of machines Pa ge 52 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 with their members’ works. One of such copyright infringements involves the action brought against Suno Inc., the owners of Suno AI and Uncharted Labs, Inc., the developer of Udio AI. RIAA alleged mass infringement of copyrighted sound recordings of its members without any permission through the training, development and operations of Suno AI and Udio AI. Copyright violations is at the heart of AI’s damaging effect on the creative entertainment industry, which also has its impact on competition laws. The procecess of training AI models with copyrighted dataset, without authorizations should be a serious concern to the entire industry stakeholders. Deploying technology to reap from where one has not sown is simply cyber-stealing, which also misleads unsuspecting and undiscerning members of the public, and this falls under the preview of unfair competition. Once a gen AI has cleared all the legal issues around copyright infringements, the next legal dynamics is the effective deployment of this AI model within the entertainment industry. Effective deployment speaks of AI becoming a tool for the human creators within the entertainment industry. Leveraging on gen AI as an effective tool that reduces costs and timing of production, and optimizing the production processes. Combining that with human’s creativity, originality and ingenuity is what will take the entertainment industry to the next level or even greater heights. For example, Jumpcut AI, helps producers swiftly read through 50,000+ scripts received from scriptwriters, and generates two-page reports for producers and directors, breaking them down into genre, subgenre, characters and similar titles. This saves enormous time, and infuses accuracy into the whole process, avoiding the consequences of human fatigue that potentially comes with manual selection process of over 50,000 scripts. Furthermore, as we have discussed above, to enjoy maximum copyright protection, which includes full exclusive derivative rights over the “original works of authorship”, human creativity must be the “finishing line” or “final expression” for the AI generated works. This aligns with theory that when AI determines the expressive elements of an output material, the generated output material does not qualify as human authorship. Therefore, its human authorship that infuses the needed creativity and originality that qualifies the work for “original works of authorship”, which makes it copyright protectable, and gives it full exclusive right to derivative exploitation. This keeps the industry profitable, lucrative and attractive to investors, thus, leveraging on the full benefits of gen AI. By this, gen AI will still be acceptable, and not seen as a “threat” to humans, but rather a powerful tool for effectiveness and maximum returns for the entertainment industry. Particularly, developing economies in Africa like Nigeria, South Africa, Kenya, Ghana, and other developing economies like India, Brazil, Turkey, Mexico, Peru, etc, that have achieved massive growth in their entertainment industry, can leverage on gen AI to further scale and globalize their local entertainment industry. With effective IP protection laws, gen AI deployed alongside the existing abundant human creativity, those countries are bound to witness a monumental leap in the quality and volume of creative works that can qualify as “original works of authorship.” Doshi & Hauser (2024) conducted a research on the effectiveness of gen AI to enhance creativity (novelty) in script development. It was found that gen AI assistance increases both the novelty and usefulness of stories more than baseline human efforts without gen AI assistance. The research confirmed that access to gen AI idea leads to greater creativity and novelty in writing. This affirms the position that deploying gen AI as a tool within the entertainment industry could lead to enhanced creativity and novelty, further expanding the value within the industry. What this implies is that with AI assistance, “original works of authorship” can still be developed, protected under copyright law, and the derivatives fully extracted to the benefit of the industry. The industry can still remain lucrative, profitable and investor friendly; rather than being destroyed by AI; its built by AI. Finally, there are several current policies and legal frameworks that can make gen AI an effective and transformative tool within the entertainment industry, rather than becoming an existential threat to that industry. Some of these are: Regulations Several jurisdictions are still not clear on how to respond to the gen AI phenomenon. Many of them are confused on the response to give to this wave. Most especially, the fear of not intentionally stifling innovation is limiting conscious efforts to regulate these AI boom. A few jurisdictions seems to be wielding more regulatory enforcements, whilst others are just “watching”, allowing the cases to play out insides the courts. Whatever side, there has to be a definitive legal position on gen AI, to create certainty within the entertainment industry. Legal uncertainty is dangerous for any society, and this is what makes laws, including regulations, dynamic. Regulations or AI legal frameworks on the development and usage of AI models need to be set up and enforced for a much healthier and safer society. Of course, regulations that do not stifle innovation and creativity. Thus far, without such regulations, the global community has witnessed ethical issues and abuses, such as massive copyright and trademark infringements and violations, competition abuses, misleading and defrauding members of the public with fake celebrity audios and videos, and other dangerous tendencies, such as blasphemous and scandalous fake voice-overs and deep videos. The governments must find a way of stepping into this global “chaos” through effective regulations, and developing policies around the development and usage of AI technologies. Otherwise, if left unregulated or unpoliced, the emergence and explosion of AIs would develop into an uncontrollable phenomenon that is inimical to humanity, most especially, the potential damaging effect on the creative industry. Gen Pa ge 53 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 AI should enhance human development and evolution, and not destroy its existence. Why should an AI generated song that uses the vocals of popular artist(s) to deceive unsuspecting members of the public, ultimately, generate commercial benefits to the AI owners? This is absolutely unethical and fraudulent, which should be prevented or brought under control through well thought-out and effective regulations. Strengthening Current Copyright Laws Emanating from regulation requirement is strengthening and enforcement of copyright laws to prevent the mass stealing of copyrighted materials by new and emerging AI models. The strengthening and enforcement of copyright laws should establish a balance between fair use, transformative use, and de minimus on one hand and the author’s derivative rights on the other hand. This is extremely essential for the survival and continuous growth of the entertainment industry. Concepts like copyright protections for authors of publications that describe tribal and community heroes, human legends that had ever lived; protecting them against unauthorized use of their publications to generate new videos, images, audios, or any form of multimedia. This copyright protection should have built into it compensation structure from the gen AI companies, for the authors or even custodians of those heroes/legends stories, or even the host communities. Not all these heroes or legends belong to public domains. Furthermore, trademark, competition, and tort laws could be deployed in response to malicious use of gen AI technologies. A resounding legal framework specifically for gen AI will be an exciting thing for the entertainment industry, which will definitely enhance productivity, utility, and creativity, rather than destroying creativity or even humanity within the entertainment industry. Developing Solid Business Models Around AIs that Benefit All The emergence and adoption of gen AI within the entertainment industry has led to the emergence of new business models within the industry. As disruptive as these developments might be, gen AI developers/owners and the copyright owners, whose datasets are required to train this models, could ultimately benefit from well- crafted business models. Different business models are emerging, and one of them is the commissioning of known celebrities to create digital characters, by big studios and brands. This is gradually triggering the emergence of marketplaces for digital characters and other digital paraphernalia of motion pictures production. Other potential business models are subscription and licensing models, where potential gen AI developers could enter into contracts with association of copyright owners to licence the original works of their members in bulk. This could be an upfront licensing model or a revenue sharing model, which constitutes a win-win situation for everyone, and completely scales gen AI’s impact within the entertainment industry. Teach WWW3.0 at Schools AI technologies should be nurtured and taught at all schools’ levels, most especially in developing countries, bringing the generality of humans into AI’s development and evolution. Furthermore, law schools should increase the adoption of the teaching of AI technologies as part of law schools’ curriculum, to enhance better understanding of the technology and the importance of formulation and enforcement of legal frameworks that can protect everyone. REFERENCES Barco, M. de. (2023, June 14). It’s gonna be a hot labor summer’ — Unionized workers show up for striking writers. NPR. https://www.npr.org/2023/06/14/1181947862/ writers-strike-union-solidarity Belcic, I., & Stryker, C. (n.d.). What is supervised learning? IBM. https://www.ibm.com/think/topics/ supervised-learning Bergmann, D. (2023). What is semi-supervised learning? IBM. https://www.ibm.com/think/topics/semi- supervised-learning Blaszczyk, M. (2023). Impossibility of emergent works’ protection in U.S. and EU copyright law. North Carolina Journal of Law & Technology, 25, 1. https://papers.ssrn. com/sol3/papers.cfm?abstract_id=4519511 Burrow-Giles Lithographic Co. v. Sarony, 111 U.S. 53 (1884). See also, Urantia Foundation v. Maaherra, 114 F.3d 955 (9th Cir. 1997). Compendium of U.S. Copyright Office Practices (3d ed. 2021), § 306, 54. U.S. Copyright Office. https://www. copyright.gov/comp3/docs/compendium.pdf Copyrightable authorship: What can be registered (n.d.). U.S. Copyright Office. https://www.copyright.gov/ comp3/chap300/ch300-copyrightable-authorship.pdf Copyright registration guidance: Works containing material generated by artificial intelligence. (2023). Federal Register, 88(51). https://www.govinfo.gov/ content/pkg/FR-2023-03-16/pdf/2023-05321. pdf#page=3 Creative economy contributes over $1.1 trillion to the U.S. economy. (2024). National Assembly of State Arts Agencies. https://nasaa-arts.org/communication/ creative-economy-contributes-over-1-1-trillion-to- the-u-s-economy/#:~:text=These%20data%2C%20 published%20by%20the,added%20to%20the%20 U.S.%20economy Doshi, A. R., & Hauser, O. P. (2024). Generative AI enhances individual creativity but reduces the collective diversity of novel content. Science Advances, 10(28). https://doi.org/10.1126/sciadv.adn5290 Generative artificial intelligence and copyright law. (2023). Congressional Research Service. https://crsreports. congress.gov/product/pdf/LSB/LSB10922 Historical and revision notes to the Copyright Act: House Report No. 94–1476. (n.d.). Pa ge 54 https://journals.e-palli.com/home/index.php/ajsl Am. J. Soc. L. 4(1) 46-54, 2025 How does AI actually work? (2021). CSU Global Blog. https://csuglobal.edu/blog/how-does-ai-actually-work How much does it cost to make a movie? Everything you need to know. (n.d.). Nashville Film Institute. https:// www.nfi.edu/how-much-does-it-cost-to-make-a- movie/#:~:text=Introduction%20to%20Film%20 Budgeting,that%20are%20subject%20to%20change IBM. (n.d.). What is machine learning? https://www.ibm. com/think/topics/machine-learning IBM. (2021). What is unsupervised learning? https://www. ibm.com/think/topics/unsupervised-learning Ingham, T. (2023). Universal Music Group responds to ‘fake Drake’ AI track. Music Business Worldwide. https:// www.musicbusinessworldwide.com/universal-music- group-responds-to-fake-drake-ai-track-streaming- platforms-have-a-fundamental-responsibility/ Kinder, M. (2024). Generative AI: What is at stake for Hollywood writers. Brookings. https://www.brookings. edu/articles/hollywood-writers-went-on-strike-to- protect-their-livelihoods-from-generative-ai-their- remarkable-victory-matters-for-all-workers/ Korn, J. (2023, January 17). Getty Images suing the makers of popular AI art tool for allegedly stealing photos. CNN Business. https://www.cnn.com/2023/01/17/tech/ getty-images-stability-ai-lawsuit/index.html Lan, Z., Yang, S., Fan, R., Zhao, B., & Yan, Y. (2023). Innovation or piracy? Empirically demarcating AI painting copyright infringement boundary. https:// www.at lant is-press.com/proceedings/pmis- 23/125989284 Marcolivio, M. (2024). AI and the movies: A blockbuster success or a big budget disaster? IMD. https://www.imd. org/ibyimd/artificial-intelligence/ai-and-the-movies- a-blockbuster-success-or-a-big-budget-disaster/ McCarthy, J. (2007). What is artificial intelligence? Stanford University. https://www-formal.stanford.edu/jmc/ whatisai.pdf Ministry of Culture, Sports, and Tourism / Korea Copyright Commission & Korean Copyright Commission. (2024). A guide on generative AI and copyright. https://www.mcst.go.kr/english/policy/ pressView.jsp?pSeq=391 Movie budgets. (n.d.). The Numbers. https://www.the- numbers.com/movie/budgets/all Murel, J., & Kavlakoglu, E. (2024). What is transfer learning? IBM. https://www.ibm.com/think/topics/transfer- learning PwC. (2024). PwC global entertainment & media outlook 2024– 28. https://www.pwc.com/gx/en/news-room/ press-releases/2024/pwc-global-entertainment-and- media-outlook-2024-28.html Record companies bring landmark cases for responsible AI against Suno and Udio in Boston and New York federal courts, respectively. (2024). RIAA. https:// www.riaa.com/record-companies-bring-landmark- cases-for-responsible-ai-againstsuno-and-udio-in- boston-and-new-york-federal-courts-respectively/ Sakoui, A. (2023, July 13). Hollywood actors’ union SAG- AFTRA votes to strike. ‘We are the victims here’. Los Angeles Times. https://www.latimes.com/ entertainment-arts/business/story/2023-07-13/sag- aftra-board-strike-plan-vote Snow, S. (2023, May 5). AI writing is the opposite of thought leadership. Forbes. https://www.forbes.com/sites/ shanesnow/2023/05/05/ai-writing-is-the-opposite- of-thought-leadership/?sh=b86faf19eafa South Africa Copyright Act, 1978. Stephen Thaler v. Shira Perlmutter and The United States Copyright Office, No. 1:22-cv-01564 (D.D.C. 2022). Stryker, C., & Kavlakoglu, E. (2024). What is artificial intelligence? IBM. https://www.ibm.com/topics/ artificial-intelligence The Difference Between Labeled and Unlabeled Data. (2023). Toloka Team. https://toloka.ai/blog/labelled- data-vs-unlabelled-data Turing, A. M. (1950). Computing machinery and intelligence. Mind, 49, 433–460. https://courses. cs.umbc.edu/471/papers/turing.pdf U.S. Const. art. I, § 8, cl. 8. United Kingdom Copyright, Designs and Patents Act 1988, § 9(3). Wang, Y., & Zhang, J. (2024). Beijing Internet Court grants copyright to AI-generated image for the first time. Kluwer Copyright Blog. https://copyrightblog.kluweriplaw. com/2024/02/02/beijing-internet-court-grants- copyright-to-ai-generated-image-for-the-first-time/