Litcius/Paper detail

Pre-made Empowering Artificial Intelligence and ChatGPT: The Growing Importance of Human AI-Experts

Maria Virvou, George A. Tsihrintzis

202320 citationsDOI

Abstract

This paper investigates the augmented responsibility of human Artificial Intelligence experts in the era of empowered pre-made Artificial Intelligence (AI). The responsible and ethical use of pre-made AI is of paramount importance in this evolving technology. AI systems have the potential to impact numerous aspects of society, ranging from healthcare and finance to education and IoT. The decisions made by AI algorithms can have significant consequences for individuals, communities, and even entire industries. Using a comparison to the way widely available medicines require a prescription from medical doctors, human AI experts assume the role of evaluating, recommending, and overseeing the implementation of AI systems, even when pre-built AI solutions may seem user-friendly on the surface. The paper has explored the expanded responsibilities of human AI experts within two contemporary scenarios involving pre-made AI, encompassing LLMs and ChatGPT. These AI technologies are applied in two principal manners: initially, as standalone AI products readily accessible to a wide audience, and secondly, as elements undergoing exploration for integration into other AI-driven software and Intelligent Information Systems (IIS), with the goal of enhancing natural language processing (NLP) features within user interfaces. In all cases, the expertise of human AI professionals is indispensable, and their role is augmented. These professionals bear an increased responsibility for ensuring the responsible and ethical deployment of AI technologies, with a focus on human-centered design, bias mitigation, validation and accuracy estimation of the results, transparency promotion, and the necessary balance between automation and human oversight. This paper performs a review on pre-made AI and ChatGPT together with custom-based AI and shows that recent advance require an augmented role of human AI experts

Topics & Concepts

Transparency (behavior)Computer scienceApplications of artificial intelligenceSoftware deploymentArtificial intelligenceAutomationPrincipal (computer security)Data scienceKnowledge managementEngineeringSoftware engineeringComputer securityMechanical engineeringArtificial Intelligence in Healthcare and EducationExplainable Artificial Intelligence (XAI)Ethics and Social Impacts of AI