Generative AI
By Stefan Feuerriegel, Jochen Hartmann, Christian Janiesch, Patrick Zschech. Business & Information Systems Engineering, 2023
1,478 citations recorded by the source corpus.
Recent advancements in generative artificial intelligence (AI) have made it possible for machines to independently produce a variety of creative content. In the context of producing creative content, this essay examines the developments, difficulties, and ethical issues relating to generative AI. It looks into how generative models, such Generative Adversarial Networks (GANs) and Variational Auto encoders (VAEs), can produce realistic artwork like music, literature, and visuals. However, it is frequently discovered that GAN training is extremely unstable and frequently experiences non-convergence, mode collapse, and hyperparameter sensitivity [1]. The technical details of developing and optimizing generative models to produce desired results are covered in detail in this work. It also looks at the difficulties in guaranteeing the variety, creativity, and coherence of generated content. Additionally, the use of generative AI in the creation of original material raises ethical questions. Included in this are concerns about intellectual property, plagiarism, and possible effects on the creative industries. In specifically, the article explores the consequences of employing generative AI for content production in terms of authorship, human creativity, and the possible disruption of traditional creative practices. It also covers issues with fairness, bias, and appropriate application of generative models.
Subfields: Artificial Intelligence, Information Systems and Management, Safety Research
View Paper: https://www.semanticscholar.org/paper/ac675900f7c6c14c8488e09a2a6e8525bcf9d45a
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Maanak Gupta
- Lopamudra Praharaj
- Charankumar Akiri
- Kshitiz Aryal
- Elisabeth Parker
Citations in the corpus
- From ChatGPT to ThreatGPT: Impact of Generative AI in Cybersecurity and Privacy (2023, background): “Generative AI models are developed using deep neural networks to learn the pattern and structure of big training corpus to generate similar new content [2].”
- Ethical considerations of Generative AI (neutral): “As mentioned in McKinsey[23], GPT-3 was trained on approximately 54 terabytes of text data at a significant cost, estimated to be several million dollars.”
- Agentic AI: Autonomous Intelligence for Complex Goals—A Comprehensive Survey (2025, neutral): “The first is how Generative AI [8] works.”