Aug 2026· Current Opinion in Psychology· Vol 72, pp.
102397
· 0 citations· 54 references
Medicine
TL;DR
This review advances three claims: first, AI anthropomorphism operates at two levels: design-based manipulations that companies can implement to make their technologies more humanlike, and individual tendencies to anthropomorphize those technologies.
Abstract
Anthropomorphism has received substantial attention in research on artificial intelligence (AI) because imbuing AI-driven technologies, such as algorithms, chatbots, and embodied robots, with humanlike qualities can mitigate AI aversion, a major barrier to AI acceptance and adoption. However, anthropomorphism may also weaken advantages associated with their nonhuman nature. This review advances three claims. First, AI anthropomorphism operates at two levels: design-based manipulations that companies can implement to make their technologies more humanlike, and individual tendencies to anthropomorphize those technologies. Second, while anthropomorphized AI may sometimes be evaluated similarly to human agents, people still distinguish AI from humans in many situations. Third, anthropomorphism can produce both positive and negative outcomes depending on the activated schemas and expectations. We conclude by outlining future research on how anthropomorphized AI may reshape our understanding of humanness.
People have long speculated about the potential dangers of powerful, self-improving artificial intelligence. Much of this speculation is anthropomorphic, assuming that AI systems will behave very similarly to humans. Omohundro’s Basic AI Drives and Bostrom’s orthogonality and instrumental convergence theses are widely accepted as foundational to emerging AI risk frameworks. However, current frontier AI models—large language models (LLMs) and related architectures—possess mindware fundamentally different from that of humans, and a different value and goal structure than either Omohundro or Bostrom assumed. In particular, frontier LLMs lack a primary terminal goal—which was assumed to be the driver of an AI’s development of instrumental values and goals, and of takeover of human affairs—and instead serve as conduits for the transient goals of many organizations and individual users. Do these key differences mean that AI systems cannot develop autonomous instrumental agency, or acquire a large degree of control over human affairs? I introduce the
instrumental succession thesis
: that human controllers of powerful AI systems pursue, on the AI’s behalf, a set of instrumental dispositions that progressively increase the AI’s capabilities and lead to the AI exercising an increasing share of oversight and control over key decisions and processes, resulting in the gradual and possibly complete transfer of the locus of agency from humans to AI. This framing presents a very different perspective on AI risk and control from classic instrumental convergence, and suggests a different set of policy and technical responses, including the active pursuit of continued human–AI merger as a hedge against both extinction and irrelevance.
Preston W. Estep· Frontiers in Psychology· 0 citations
Rapid advances in robotics and artificial intelligence (AI) have intensified claims that machines are approaching, matching, or replacing human capabilities. Such claims often obscure a critical conceptual distinction: robotics, AI, and human intelligence are not equivalent forms of agency competing on a single scale. Robotics concerns embodied systems capable of sensing and acting in the physical world; AI concerns computational systems that generate predictions, recommendations, decisions, or content; and human intelligence is a biologically embodied, socially situated capacity that integrates cognition, emotion, experience, values, and accountability. This paper critically compares the three domains using recent research published from 2020 onward. It examines performance, adaptability, embodiment, creativity, decision-making, labor, trust, and ethics. The evidence demonstrates that robots and AI can exceed humans in speed, consistency, data processing, and narrowly specified tasks, while humans remain comparatively strong in contextual reasoning, open-ended adaptation, social meaning, moral responsibility, and the interpretation of ambiguous goals. The paper further argues that the dominant replacement narrative is analytically weak because real systems increasingly combine these capacities. Human–AI and human–robot collaboration can produce significant augmentation, but collaboration is not inherently synergistic and may create new risks through automation bias, opacity, deskilling, and accountability gaps. The most defensible future model is therefore neither human supremacy nor technological substitution, but deliberately governed hybrid intelligence in which task allocation reflects the comparative strengths and limitations of human and machine agents.
Ozoya Ohilebo· International Journal of Mul...· 0 citations
The paper argues that practical imitation may bypass this barrier by relying on belief and perceived equivalence rather than authentic internalization rather than authentic internalization, and may help ensure that AI remains an auxiliary tool rather than becoming a governing influence over human thought and action.
The divergence between embodied AI (AI) and human cognition arises from differences in substrate, architecture, and developmental trajectories. Human intelligence emerges from carbon-based biochemical and evolutionary processes, while embodied AI develops through non-carbon substrates and distinct control systems. This contrast raises key questions: will embodied AI follow a trajectory toward Artificial general intelligence (AGI) comparable to that of carbon-based beings, or will it diverge into novel forms of intelligence? And if divergence is inevitable, what social and ethical consequences follow? Embodied AI may increasingly replicate three core cognitive stages that we propose—perception (“I perceive, therefore I exist”), reflexivity (“I reflect on what I perceive”), and Theory of Mind (“I consider what others perceive”). These advances point toward the possibility of Artificial general intelligence (AGI), but also heighten the risk of anthropomorphizing such systems, mistaking opaque intentions for human-like cognition. To look into this, we propose the metaethics of substrate framework, which holds that the ethical challenges and responsibilities associated with intelligence cannot be reduced to abstract measures of functional performance; here, we assume that they must also account for the material substrate through which cognition emerges. In this regard, Martha Nussbaum’s capabilities approach reminds us that even biological intelligence is plural and diverse; by extension, divergence between biological-based and non-biological embodied AI may prove even greater. Recognizing these differences is vital in the current race to AGI, where market pressures often eclipse ethical reflection. Therefore, this work argues that bridging embodied AI research with frameworks such as metaethics of substrate and the capabilities approach is essential for clarifying how these systems should be situated in society, what forms of agency they may possess, and how their divergence from human cognition can be managed responsibly. es and initials are correct.
Jorge Luis Morton, Bernardo Bolaños Guerra, D. Romero· AI and Ethics· 0 citations
This work conducts semi-structured interviews with industry practitioners working with agentic AI systems, indicating that agentic AI systems are mainly explained through organizational and anthropomorphic source domains, such as employees, teams, or assistants, which embed abstract system qualities within familiar social structures.
Felix Stundzig, Vincent Heimburg, Manuel Wiesche· Proceedings of Mensch und Co...· 0 citations
The Ordering of Human-Artificial Intelligent (AI) Constructs symposium in Volume 8(2) of Law, Technology and Humans does one simple, yet entirely immodest, thing. It is trying to shift the frame of how law, humans and AI is being understood. This introduction argues that contemporary AI-law discourses rest upon particular imaginaries of order and legal ordering. It outlines these dominant imaginaries as well as why the framework-shifting work is necessary. And it showcases how more relational, creative, and socio-technically grounded understandings of AI-human constructs allow for new ways of understanding what order and ordering is and can be, as well as the many possibilities such new imaginaries open for reimagining legal order(ing) in the present-future.
M. Arvidsson, Kieran Tranter· Law, Technology and Humans· 0 citations
