1 | Term | Definition(s) | Benefits | Drawbacks | Source(s) |
|---|---|---|---|---|---|
2 | Artificial agent | an interactive, autonomous, adaptable entity instantiated at least partly on non-biological substrates | - grants capacity for taking independent, self-controlled actions to at least partly non-biological entities - clearly identifies that the entity is at least partly non-biological - addresses relationality - could be used as a general term for a class of entities built (at least initially) by humans | - may limit moral consideration because of the emphasis on agency - relationality is not clearly linked to moral consideration - may limit humans’ conceptions of the capacities of these entities - subject to concerns about “fakeness” | Floridi & Sanders, 2004 |
3 | Artificial being | a distinct and independent existence instantiated at least partly on non-biological substrates | - easily understandable - familiar - implies potential for experiential capacities based on “life-like” qualities of “being” - clearly identifies that the entity is at least partly non-biological | - may have different meanings across contexts - closely related to legal concepts like “artificial person” - colloquial - subject to concerns about “fakeness” - may imply requirement of a soul or metaphysical presence | - |
4 | Artificial consciousness | - Graziano: “a machine that contains a rich internal model of what consciousness is, attributes that property of consciousness to itself and to the people it interacts with, and uses that attribution to make predictions about human behavior. Such a machine would ‘believe’ it is conscious and act like it is conscious, in the same sense that the human machine believes and acts” - Reggia: “computational models of various aspects of the conscious mind, either with software on computers or in physical robotic devices” | - implies capacity for introspection, self-awareness, and consciousness in a human-built (at least initially) entity akin to that present in humans - emphasizes internal experiences - established and already used in scholarship | - subject to the same philosophical and empirical criticisms about “consciousness” - unclear what material components (e.g., body) are necessary - complex relationship to moral consideration - potential association with “fake” or “unnatural” - requires evidence of consciousness - may be controversial in use with the general public - difficult to observe and verify from a first person perspective | Graziano, 2017; Reggia, 2013 |
5 | Artificial entity | a distinct and independent existence instantiated at least partly on non-biological substrates | - easily understandable - familiar - reflects diversity in possible types of entities - clearly identifies that the entity is at least partly non-biological - not tied to existence of a soul or metaphysical presence | - abstract - may have different meanings across contexts - used commonly to refer to corporations - subject to questions about “fakeness” - may prompt objectification | - |
6 | Artificial general intelligence | “a software program that can solve a variety of complex problems in a variety of different domains, and that controls itself autonomously, with its own thoughts, worries, feelings, strengths, weaknesses and predispositions” | - established and commonly used - may prompt increased similarity to humanity and thus make it easier to promote moral consideration - emphasizes internal phenomena (e.g., software programming, mental capacities) rather than external properties (e.g., material structure, embodiment) | - implicit connection with human-like or a human-like basis for intelligence requires humans to be the initial standard - potential association with “fake” or “unnatural” - limited by existing conceptual bounds of AIs as cognitively skilled machines - definition susceptible to meaning changes over time as technology and conceptions of “intelligence” change - may require evidence of self-awareness or consciousness | Goertzel & Pennachin, 2007 |
7 | Artificial intelligence | “systems that can decide what to do and do it” and that vary based on how human-like and/or rational their thought processes and/or behavior are | - well-known and commonly used - entails specific capacities (e.g., memory, learning, reasoning) | - used interchangeably to signify the field of study, an individual entity, or a network of entities - potential association with “fake” or “unnatural” - established usage is conflated with sophisticated cognition - depends on conceptions of human-like intelligence - definition susceptible to meaning changes over time as technology and conceptions of “intelligence” change - not typically connected with moral consideration - may be prone to hype | Russell & Norvig, 1995 |
8 | Artificial moral agent | - Cervantes et al.: “artificial agents capable of making ethical and moral decisions” - Himma: “an [artificial] agent with the capacities to choose its actions ‘freely’ and understand the basic concepts and requirements of morality, capacities that also presuppose consciousness” | - promotes association of artificial entities with morality - clearly signifies an individual - entails specific types of decisions and actions - considers internalexperiences of artificial entities | - may be associated only with taking moral action rather than with receiving moral consideration - may require evidence of consciousness - context specific moral boundaries may introduce safety concerns or present a threat to humans | Cervantes et al., 2020; Himma, 2009 |
9 | Artificial sentience | - artificial entities with the capacity for positive and negative experiences, such as happiness and suffering - the capacity for positive and negative experiences manifested in artificial entities | - clear connection to moral consideration from the nonhuman animal research on sentience - represents specific, valenced experiential capacities - does not require human-like intelligence - dissociable from cognitive capacities like problem-solving and analytical thinking - may be more distinctive than “mind” or “consciousness” - prioritizes perception, emotion, and behavior | - could signify an individual entity, a collection of entities, a psychological/phenomenal experience, or a field of study - little established scholarly or applied usage - potential association with “fake” or “unnatural” - potential conflation with “artificial consciousness” due to historical philosophical link between “sentience” and “consciousness” - difficult to observe and verify from a first person perspective | Sentience Institute, 2021 |
10 | Digital mind | - S&B: “machine minds with conscious experiences, desires, and capacity for reasoning and autonomous decision-making” - Sotala: “a mind that runs on a computer” | - bridges interdisciplinary conceptions of “mind” (e.g., psychology, artificial intelligence) - inclusive of a range of machine-based entities (e.g., whole brain emulations, AI algorithms) - existing theorizing on moral status and the potential to experience suffering - highlights non-material aspects of entities | - some research uses this term to refer to human brains interacting with digital media - tied to instrumental use purposes (e.g., “digital mind mapping,” “digital mind games”) - implied connection with the human brain or systems initially modelled on biological brains | Shulman & Bostrom, 2021; Sotala, 2012 |
11 | Digital person | - EA: “a person running on digital computing hardware” - Karnofsky: digital copies of humans such as simulated human brains and digital descendants of humanity | - connection to “humanness” and human capacities like sentience that may increase moral consideration - framework may make the concept more accessible to the general public - highlights non-material aspects of entities | - implied connection with the human brain or systems initially modelled on biological brains - excludes many types of artificial entities, notably those who are less human-like or not descended from humans - the future framing of the concept may reduce its near- and medium-term impact | Effective Altruism Forum, 2021; Karnofsky, 2021 |
12 | Full ethical agent | “can make explicit ethical judgments and generally is competent to reasonably justify them” | - implies specific and directional action tied to morality - can be used in connection with making moral judgments - implies greater depth of agency with the term “full” - can apply to any entity with the capacity to make ethical judgments, human or nonhuman | - does not uniquely distinguish artificial entities and their capacities - does not consider the perceptual, affective, and experiential capacities of the entity - not commonly used - may promote a false binary distinction between moral agency and moral patiency | Moor, 2006 |