Security, Privacy, and Ethical Challenges of Artificial Intelligence in Large Language Model Scope: A Comprehensive Survey
Tuyen T. Nguyen, Huyen T. T. Vu, Hoa Ngoc Nguyen · 2024
Artificial intelligence (AI) models like ChatGPT and LLama have changed how we understand and create human-like language. They understand language well, can generate text like a person, know the context, and solve problems effectively. This makes them useful in many areas like search engines, customer service, and translation. Recently, these models have also caught the attention of the security field, uncovering weaknesses in security and proving useful for security tasks. This paper offers an extensive review of the security, privacy, and ethical issues associated with AI, focusing on LLMs and their impact on various domains. We explore the vulnerabilities of LLMs, propose poten-tial defense mechanisms, and highlight the broader implications of AI on human society. This survey intends to offer a clear perspective for researchers and practitioners, and policymakers on responsibly developing and deploying AI technologies.