Unsealed files describe OpenAI "research mission" to supplant human authors (www.publishersweekly.com)

Unsealed court documents in the ongoing copyright infringement case between the Authors Guild and OpenAI/Microsoft reveal that OpenAI executives acknowledged their training of AI models on copyrighted texts, including books, and anticipated substantial market displacement of human authors. Notably, a 2019 internal note admits to using "pirated stuff" for training GPT-3, highlighting a troubling disregard for copyright compliance. Additionally, OpenAI's policy director Jack Clark noted in 2020 that their advancements with future models would likely instigate fear among genre fiction authors regarding job security while indicating a willingness to ignore such concerns. The implications of these revelations are profound for the AI/ML community, raising ethical questions about intellectual property and the potential for AI systems to replace human creators. Tarun Gogineni, leading OpenAI’s writing quality improvements, described his research mission as one aimed at supplanting human authors, dismissing industry complaints as "acceptable economic disruption." With these statements, the tension between AI innovation and traditional authorship intensifies, marking a significant moment in the debate over the role of AI in creative industries and the vital need for robust discussions around copyright and ethical AI practices.