Large language models (LLMs) and dialogue agents have existed for years, but the release of recent GPT models has been a watershed moment for artificial intelligence (AI) research and society at large. Immediately recognized for its generative capabilities and versatility, ChatGPT's impressive proficiency across technical and creative domains led to its widespread adoption. While society grapples with the emerging cultural impacts of ChatGPT, critiques of ChatGPT's impact within the machine learning community have coalesced around its performance or other conventional Responsible AI evaluations relating to bias, toxicity, and 'hallucination.' We argue that these latter critiques draw heavily on a particular conceptualization of the 'human-centered' framework, which tends to cast atomized individuals as the key recipients of both the benefits and detriments of technology. In this article, we direct attention to another dimension of LLMs and dialogue agents' impact: their effect on social groups, institutions, and accompanying norms and practices. By illustrating ChatGPT's social impact through three disruptive events, we challenge individualistic approaches in AI development and contribute to ongoing debates around the ethical and responsible implementation of AI systems. We hope this effort will call attention to more comprehensive and longitudinal evaluation tools and compel technologists to go beyond human-centered thinking and ground their efforts through social-centered AI.

翻译：暂无翻译