Author: Lakhtakia, Akhlesh; Lakhtakia, Natalya S.
Description: Artificial intelligence (AI) is rapidly moving from a specialized research topic into a widely deployed general-purpose technology, driven by advances in transformer-based architectures, convincing enough productivity gains, large private investment, geopolitical competition, and expanding educational adoption. AI comprises two distinct sets of methodologies: (i) predictive AI, which is grounded in statistical inference and aims to model measurable physical reality, and (ii) generative AI, which produces plausible synthetic content without intrinsic verification against physical reality. Whereas predictive AI can strengthen human decision-making through rigorously tested models grounded in impeccable mathematics, generative AI introduces systemic risks that extend beyond technical reliability. Generative AI is a wicked problem, because its consequences are deeply entangled across environmental, human, ethical, educational, and cultural domains. Generative AI accelerates energy demand, greenhouse-gas emissions, water consumption, and electronic waste, thereby amplifying environmental burdens. It also intensifies inequality through an expanding AI divide, undermines intellectual property norms, and relies on poorly compensated human labor for ongoing data annotation and content moderation. Furthermore, hallucinated outputs, misinformation, and AI slop degrade epistemic trust in online information ecosystems. Ethical dangers include privacy erosion, algorithmic bias in societal decision-making, and increased risk of catastrophic failures in high-stakes contexts such as policing and warfare. Generative AI may reduce critical thinking in students through cognitive offloading and weaken originality and writing skills. AI may serve as a useful servant to specialists for workflow acceleration but we must reject it for general-purpose use, with strict regulation necessary for public chatbot deployment to preserve human judgment and sustain civilization.
Subject headings: Agentic AI;Â AI hallucination;Â AI psychosis;Â Wicked problem;Â Generative AI;Â Predictive AI;Â Large language models;Â AI divide; Artificial intelligence
Publication year: 2026
Journal or book title: Biologically Inspired Materials, Processes, and Systems (BIMPS) 2026
Volume: 13944
Pages: 80-96
Find the full text: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/13944/139440A/Is-artificial-intelligence-a-good-servant-but-a-bad-master/10.1117/12.3077874.short
Find more like this one (cited by): https://scholar.google.com/scholar?cites=3387769333721612207&as_sdt=1000005&sciodt=0,16&hl=en&scioq=Illusions+of+Understanding+from+Outsourcing+Thinking+to+LLMs
Serial number: 4224