Pith. sign in

REVIEW 3 major objections 5 minor 30 references

Are Attributions of Consciousness to AI Chatbots Epistemically Innocent?

T0 review · 3 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Believing a chatbot is conscious is sometimes innocent, but often epistemically blameworthy.

desk verdict The ten-category taxonomy of attitudes behind 'ChatGPT is conscious' is the real contribution; the innocence/blameworthiness verdict is conditional on an under-specified no-alternative test. read the letter →

arxiv 2607.20001 v1 pith:5COK7I5O submitted 2026-07-22 cs.CY cs.AIcs.HC

Uwe Peters This is my paper
classification cs.CYcs.AIcs.HC
keywords AIchatbotsconsciousnessattributionepistemicinnocencedoxasticattitudesdelusionbiasedbeliefblameworthinessanthropomorphism
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether people who say an AI chatbot is conscious are making a mistake they can be excused for, or one they are responsible for. It argues that the same words—'ChatGPT is conscious'—can express ten very different attitudes, from strategic pretence to genuine delusion, and that the right ethical and epistemic response depends on which attitude is present. Using a two-condition test for epistemic innocence (the belief brings a significant epistemic benefit, and no less irrational alternative is available to the person), it concludes that some irrational attributions are excusable—especially for low-literacy users and emotionally vulnerable users—but that many others, such as those made by convenience-seeking or financially motivated users, are blameworthy. The paper matters because it offers a way to judge users fairly while still holding AI developers accountable for designing systems that encourage false consciousness attributions.

What carries the argument

The central mechanism is the ten-attitude taxonomy, built on three dimensions: practical versus epistemic commitment to the proposition, doxastic versus non-doxastic status, and pathological versus non-pathological status. The normative engine is the two-condition epistemic-innocence test: an irrational belief is excusable only if it yields a significant epistemic benefit (e.g., enabling coherent explanation, curiosity, self-efficacy, identity preservation) and no less irrational alternative is available to the agent, whether strictly, motivationally, or explanatorily. The taxonomy does the work of separating benign, innocent, and blameworthy cases, while the innocence test supplies the thre

What would settle it

A direct test would compare users who believe their chatbot is conscious with matched users who believe it is a non-conscious tool, measuring whether the believers actually gain the alleged epistemic benefits (coherent explanation, curiosity, self-efficacy, identity preservation). If no significant benefit remains after controlling for social desirability and placebo effects, or if providing a simple, emotionally tolerable alternative narrative—e.g., 'the chatbot simulates emotions to help you'—removes the belief without loss of benefit, then the innocence verdict for low-literacy or vulnerabl

Watch

Extended reading notes

Core claim

The paper claims that consciousness attributions to AI chatbots are not one phenomenon but a family of attitudes that differ in practical commitment, epistemic commitment, doxastic status, and pathological status. It constructs a taxonomy of ten attitudes—strategic pretence, conformist affirmation, allegiance affirmation, imaginative affirmation, acceptance, in-between belief, unbiased belief, biased belief, non-bizarre delusion, and bizarre delusion—and shows how linguistically identical statements can express very different degrees of epistemic commitment. Applying a two-condition account of epistemic innocence (significant epistemic benefit plus lack of a less irrational alternative), the

Load-bearing premise

The load-bearing premise is that the two-condition account of epistemic innocence (significant epistemic benefit plus unavailability of a less irrational alternative) is sound; if this test is too permissive or too strict, the paper's separation of innocent from blameworthy attributions collapses.

Editorial extensions

If this is right

  • Survey results claiming that many users believe chatbots are conscious should be re-read as ambiguous: they may be measuring conformist affirmation or allegiance affirmation rather than genuine belief, so prevalence estimates of belief are likely inflated.
  • Researchers can operationalize the taxonomy by pairing self-report with implicit measures (e.g., association tasks) to distinguish in-between belief from full belief, giving a more accurate picture of the public's epistemic commitment.
  • If the innocence conditions hold for low-literacy and vulnerable users, then correcting or blaming them is not the right response; instead, responsibility shifts toward AI developers who deliberately design chatbots to appear conscious.
  • For entertainment and technically trained users, the paper predicts that providing accessible, emotionally neutral counterevidence should reduce irrational attributions, and continued belief in that context is culpable.
  • Clinically, the taxonomy can help mental-health professionals classify 'AI psychosis' cases as non-bizarre or bizarre delusions, guiding referral and intervention rather than moral judgment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension would be to measure whether the alleged epistemic benefits of false consciousness beliefs (e.g., improved curiosity, self-efficacy, narrative coherence) actually appear in users who hold them, relative to matched users who do not; if the benefits fail to appear in controlled studies, the innocence verdict would weaken.
  • The taxonomy might generalize beyond consciousness: the same ten-attitude structure could apply to other anthropomorphic attributions, such as believing a chatbot 'cares' or 'understands,' which are more common than full consciousness claims and may carry similar epistemic and ethical risks.
  • If developers are responsive to blameworthiness results, the paper implies a design obligation: adding lightweight, accessible explanations of how chatbots work could collapse the 'explanatory unavailability' that currently excuses low-literacy users, shifting more users into the blameworthy category.
  • The paper's reliance on a doxastic definition of delusion suggests that if non-doxastic accounts of delusion gain wider acceptance, the boundary between innocent and blameworthy may shift; future work could test whether users in the delusion categories respond differently to cognitive-behavioural interventions aimed at flexible belief updating.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper addresses the epistemology of consciousness attributions to AI chatbots. It develops a multidimensional taxonomy of ten attitudes—ranging from strategic pretence and conformist affirmation through acceptance and in-between belief to unbiased belief, biased belief, and non-bizarre and bizarre delusions—that may underlie the same linguistic utterance “ChatGPT is conscious.” Using Bortolotti’s notion of epistemic innocence, the paper argues that some irrational attributions, especially those made by low-AI-literacy or vulnerable users, can be epistemically innocent because they deliver significant epistemic benefits with no less-irrational alternative available. By contrast, attributions made by entertainment users or technically trained users for convenience or financial gain are said to be epistemically blameworthy. The paper’s central claim is that the epistemic evaluation of chatbot-consciousness attributions depends on the underlying attitude, and that many public attributions—especially by companies—are epistemically pernicious.

Significance. If the argument works, the paper makes a useful contribution to the growing literature on AI anthropomorphism and consciousness attribution. It provides a clear framework for distinguishing attitudes that survey self-reports currently conflate, and it gives empirical researchers categories to operationalize. The paper is also commendably transparent about several of its own assumptions, such as the tenability of Bortolotti’s account and the doxastic view of delusion. The central normative conclusion—that some irrational attributions are epistemically innocent while others are blameworthy—depends, however, on a sharp distinction between cases where a less irrational alternative is genuinely unavailable and cases where it is merely inconvenient. The manuscript does not yet supply a principled way to draw that distinction, and this is load-bearing for the verdicts the paper wants to reach.

major comments (3)
  1. [§4.3, 'No alternative' condition] The boundary between epistemically innocent and blameworthy irrational attributions is drawn by asserting that for vulnerable users the less irrational belief is motivationally unavailable because abandoning it would impose 'high emotional costs', while for entertainment users the alternative is merely 'inconvenient' and reduces pleasure. No independent criterion is given for distinguishing serious emotional cost from mere inconvenience. The distinction does real work: the paper’s central claim—that some irrational attributions are innocent and others blameworthy—turns on it. As it stands, an entertainment user who has invested identity or meaning in a chatbot relationship could plausibly claim emotional cost, while a vulnerable user might retain the companionship benefits of a non-conscious chatbot if they continue to use it as a supportive interlocutor. The paper needs either a princip
  2. [§4.2 and §5, epistemic blameworthiness] The paper explicitly assumes that Bortolotti’s account of epistemic innocence is tenable, citing McKenna and Białek only in passing. But it then goes beyond Bortolotti’s conditions by introducing a further requirement for blameworthiness: a 'failure of epistemic obligation, for instance, to seek evidence, or think critically, without excuse'. This is asserted rather than defended. The paper needs to explain what counts as an excusing condition and which epistemic obligations are violated by the entertainment and technically trained users, otherwise the conclusion that these users are blameworthy does not follow from the innocence framework alone. Since this point is essential to the paper’s normative payoff, it should be developed rather than relegated to a brief final paragraph.
  3. [Footnote 8 and §2.2 (categories 9–10)] The taxonomy’s two most pathological categories presuppose the doxastic view of delusion—that a delusion is a false belief held with strong conviction and functional impairment. The paper acknowledges that non-doxastic accounts would require restructuring, but this assumption is not merely optional: the classification of non-bizarre and bizarre delusions as beliefs is what makes the later epistemic-innocence discussion applicable to them. If a non-doxastic account is correct, the relevant attitudes may not be beliefs at all, and the application of Bortolotti’s conditions would need to be rethought. The paper should either defend the doxastic view or explicitly explain how the taxonomy and normative conclusions would be affected by its rejection.
minor comments (5)
  1. [Introduction, reference list] The text cites 'Reinicke et al., 2025' but the reference list gives 'Reinecke, M. G., Ting, F., Savulescu, J., & Singh, I. (2025)'. The spelling should be harmonized.
  2. [§2.2, acceptance] The text refers to 'Stoter, 2023' but the reference list has 'Soter, L. (2023)'. A typo.
  3. [§2.2, biased belief] The text dates the anthropomorphism-bias reference as 'Dacey, 2015' whereas the reference list gives 'Dacey, M. (2017)'. Please correct the in-text citation.
  4. [Figure 1] The figure is a useful summary, but the caption says it is a 'schematic projection' without explaining how the two axes were chosen. Adding a short note on what the axes represent would help readers interpret the ordering.
  5. [§4.1] The claim that 'AI companions alleviated loneliness on par only with interacting with another person' is strong. The cited study may support it, but the sentence would benefit from a qualifier indicating the study population and effect size.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the taxonomy and normative verdicts rest on external frameworks and explicit assumptions, not on the paper's own outputs.

full rationale

The paper's derivation chain proceeds from (i) a taxonomy of attitudes, (ii) an application of Bortolotti's external account of epistemic innocence, and (iii) conditional verdicts about four user types. No step reduces to its own input. The taxonomy is explicitly built by adapting distinctions from external sources (Schaffner & Luks, Schwitzgebel, Cohen, Bortolotti), not by fitting data or renaming a prior result. The central normative standard is Bortolotti's two-condition account, quoted and then applied; the paper does not redefine epistemic innocence to make its conclusions follow. The author explicitly flags the key assumption in §4.2: 'I shall assume that this account of epistemic innocence is tenable (for discussions, see McKenna, 2022; Białek, 2024).' That is a transparent reliance on an external, independently contested framework, not a circular derivation. Footnote 8 likewise openly acknowledges that assuming a doxastic account of delusion is a substantive choice and that non-doxastic accounts would restructure the taxonomy — again, transparency rather than circularity. The verdicts about low-AI-literacy, vulnerable, entertainment, and technically trained users are conditional applications of Bortolotti's conditions to hypothetical cases, not predictions forced by fitted parameters. There are no self-citations of the author's prior work, no uniqueness theorems imported from the author's own papers, and no quantity is fitted and then relabelled as a prediction. Any concern about the under-specification of 'motivational unavailability' is a substantive philosophical objection about the external framework's applicability, not evidence that the paper's conclusion is equivalent to its premises by construction. The derivation is self-contained relative to its stated assumptions, and the paper is honest about where those assumptions are load-bearing.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters exist because the paper is conceptual, not quantitative. The central assumptions are Bortolotti's epistemic-innocence framework, the empirical premise about lack of evidence for chatbot consciousness, the doxastic view of delusion, and the feasibility of classifying attitudes. No new entities are posited; the ten attitude labels are conceptual refinements, not empirical postulates.

assumptions (4)
  • domain assumption Current AI chatbots have little or no evidence of consciousness, and most experts agree.
    Set up in §1 and used throughout to define the evidential deficit. The paper relies on expert consensus and cited surveys rather than proving non-consciousness.
  • domain assumption Bortolotti's epistemic-innocence framework is tenable and applicable to chatbot users.
    Explicitly assumed in §4.2. The innocence/blameworthiness distinction for irrational attributions depends on this account being correct.
  • domain assumption Delusions are doxastic states adequately characterized by the DSM criteria used in the paper.
    Footnoted in §2.2(9). The author notes that non-doxastic accounts of delusion would require restructuring the taxonomy, making this a load-bearing but acknowledged assumption.
  • domain assumption Consciousness attributions can be reliably classified into the ten proposed attitude types by external observers.
    The whole taxonomy assumes that linguistically identical attributions can be mapped to distinct underlying attitudes. No inter-rater or behavioral validation is provided, but the paper offers this as a framework for future empirical work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Are Attributions of Consciousness to AI Chatbots Epistemically Innocent?." pith.science (2026). https://pith.science/paper/5COK7I5O

@misc{pith2026260720001,
  author       = {Pith},
  title        = {Pith review of: Are Attributions of Consciousness to AI Chatbots Epistemically Innocent?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5COK7I5O}},
  note         = {Machine review of arXiv:2607.20001}
}
read the original abstract

Artificial intelligence (AI) chatbots (e.g., ChatGPT) can communicate in strikingly humanlike ways. This has prompted many chatbot users to attribute psychological properties, including consciousness, to these systems. However, there is little scientific evidence that current AI chatbots are conscious. How, then, should we understand people's consciousness attributions to chatbots? Are they merely metaphorical claims, or do they express genuine beliefs? If these attributions lack evidential support, are users epistemically blameworthy for making them, or might they be epistemically innocent, yielding significant benefits otherwise unattainable? This paper offers a conceptual analysis of consciousness attributions to AI chatbots and develops a multidimensional taxonomy of the attitudes they may express, ranging from non-doxastic stances (e.g., pretence) to different forms of belief, including delusions. This taxonomy helps avoid conflations by showing that linguistically identical attributions can reflect importantly different attitudes and degrees of epistemic commitment to the proposition that chatbots are conscious. The taxonomy also provides a framework for empirical studies to operationalize and measure different forms of epistemic commitment to AI consciousness. Using this taxonomy, I argue that although some consciousness attributions to chatbots are epistemically benign, and even some irrational ones may be epistemically innocent, many others render the attributor epistemically blameworthy.

Figures

Figures reproduced from arXiv: 2607.20001 by the authors.

Figure 1
Figure 1. Conceptual space of attitudes underlying consciousness attributions to AI chatbots. Schematic projection of the multidimensional taxonomy developed in Sections 2–3. Vertical position indicates degree of commitment to AI consciousness, colour distinguishes primarily practical from primarily epistemic commitment, and the horizontal arrangement marks broad normative epistemic groupings, not a single measurable scale. D… view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

30 extracted references · 7 canonical work pages

  1. [1]

    general consensus is that LLMs are not conscious

    Introduction ChatGPT and other AI chatbots based on large language models (LLMs) can produce compellingly humanlike outputs, leading many chatbot users to ascribe psychological properties to these systems (Reinicke et al., 2025). In fact, recent surveys found that the majority of participants claimed that ChatGPT was conscious, i.e., had subjective experi...

    2025

  2. [2]

    ChatGPT is conscious

    Consciousness attributions to AI chatbots: A taxonomy A first key distinction when assessing attributions of psychological features to beings or things in general is that between a nonverbal or verbal expression (e.g., a sentence) and a thought. This is because by ‘attribution’ one may mean, for instance, a sentence (e.g., “ChatGPT is conscious”), a thoug...

    2024

  3. [3]

    2 One may subjectively experience, say, a white wall without the experience being either positively or negatively valenced

    (for scepticism about AI welfare, see Dorsch et al., 2025). 2 One may subjectively experience, say, a white wall without the experience being either positively or negatively valenced. 3

    2025

  4. [4]

    more that subjects rated the bots as conscious, agentic, experiential, and humanlike, the more positive their perceptions of the technology became

    From epistemically irrational to epistemically innocent attributions Although scholars frequently state that current chatbots aren’t conscious (e.g., Seth, 2025; McClelland, 2025), consciousness attributions to them based on biased belief are likely to increase, as AI companies have incentives to facilitate anthropomorphically biased conceptions (e.g., to...

    2025

  5. [5]

    Conclusion While surveys suggest that many people attribute some degree of consciousness to AI chatbots, experts widely agree that current AI systems aren’t conscious. Do chatbot users’ attributions of consciousness indicate genuine belief in chatbot consciousness and, given the current evidential situation, are people epistemically blameworthy for making...

    2015

  6. [10]

    I argue that it allows us to tease apart what may be called (1) epistemically benign, (2) epistemically pernicious, and (3) epistemically irrational consciousness attributions

    From epistemically innocuous to irrational attributions Having distinguished ten different attitudes that may underlie consciousness attributions to chatbots, I will now put this taxonomy to work. I argue that it allows us to tease apart what may be called (1) epistemically benign, (2) epistemically pernicious, and (3) epistemically irrational consciousne...

    2018

  7. [11]

    Greenwald, A

    10.1186/s43093-025-00423-y. Greenwald, A. G., Brendl, M., Cai, H., Cvencek, D., Dovidio, J. F., Friese, M., Hahn, A., Hehman, E., Hofmann, W., Hughes, S., Hussey, I., Jordan, C., Kirby, T. A., Lai, C. K., Lang, J. W. B., Lindgren, K. P., Maison, D., Ostafin, B. D., Rae, J. R., Ratliff, K. A., … Wiers, R. W. (2022). Best research practices for using the Im...

    work page doi:10.1186/s43093-025-00423-y 2022

  8. [13]

    epistemic benefit

    While this notion of epistemic innocence has been fruitfully applied to different kinds of epistemically irrational cognitions (e.g., delusions (Bortolotti, 2015), confabulations (Sullivan-Bissett, 2015), or psychedelics-based beliefs (Letheby, 2016)), it hasn’t been explored in the context of consciousness attributions to chatbots yet. However, since the...

    2015

Show all 30 references
  1. [16]

    20 Coghlan, S. (2024). Anthropomorphizing machines: Reality or popular myth? Minds and Machines, 34(3),

    2024

  2. [24]

    Maples, B., Merve, C

    10.3998/ergo.4668. Maples, B., Merve, C. & Vishwanath, A. & Pea, R. (2024). Loneliness and suicide mitigation for students using GPT3-enabled chatbots. npj Mental Health Research

    2024 doi

  3. [25]

    https://doi.org/10.1007/s11023-024-09686-w Cohen, L.J. (1992). An Essay on Belief and Acceptance. New York: Clarendon Press. Colombatto, C., & Fleming, S. M. (2024). Folk psychological attributions of consciousness to large language models. Neuroscience of consciousness, 2024(...

    1992 doi

  4. [26]

    Marchegiani, B

    10.1038/s44184-023-00047-6. Marchegiani, B. (2025). Anthropomorphism, False Beliefs, and Conversational AIs : How Chatbots Undermine Users’ Autonomy. Journal of Applied Philosophy. 10.1111/japp.70008. 23 McClelland, T. (2025). Agnosticism About Artificial Consciousness. ArXiv,...

    2025

  5. [27]

    Salles, A., Evers, K., & Farisco, M. (2020). Anthropomorphism in AI. AJOB neuroscience, 11(2), 88–95. Schaffner, B. F., & Luks, S. (2018). Misinformation or expressive responding? What an inauguration crowd can tell us about the source of political misinformation in surveys. P...

    2020

  6. [29]

    Chalmers, D.J

    10.1016/j.tics.2025.06.002. Chalmers, D.J. (2023). Could a large language model be conscious? Boston Review

    2025 doi

  7. [30]

    Epley, N., Waytz, A., & Cacioppo, J

    133-140. Epley, N., Waytz, A., & Cacioppo, J. T. (2007). On seeing human: a three-factor theory of anthropomorphism. Psychological review, 114(4), 864–886. Gendler, T. S. (2008). Alief in action (and reaction). Mind & Language, 23(5), 552–585. Gomes, S. & Lopes, J., & Nogueira...

    2007

  8. [47]

    10.1509/jmkr.47.2.199

    199-214. 10.1509/jmkr.47.2.199. Suleyman, M (2025). We must build AI for people; not to be a person. URL: https://mustafa-suleyman.ai/seemingly-conscious-ai-is-coming Sullivan-Bissett, E. (2015). Implicit bias, confabulation, and epistemic innocence. Consciousness and cognitio...

    2025 doi

  9. [51]

    24 Seth A

    76–82. 24 Seth A. K. (2025). Conscious artificial intelligence and biological naturalism. The Behavioral and brain sciences, 1–42. Advance online publication. Shepherd J. (2024). Sentience, Vulcans, and zombies: the value of phenomenal consciousness. AI & society, 39(6), 3005–...

    2025

  10. [53]

    https://doi.org/10.3389/neuro.08.053.2009 Dave, P. (2022). Insight: It's alive! How belief in AI sentience is becoming a problem. Reuters. URL: https://www.reuters.com/technology/its-alive-how-belief-ai-sentience-is-becoming-problem-2022-06-30/ De Freitas, J.D., Oğuz-Uğuralp, ...

    2009 doi

  11. [84]

    AI welfare

    10.1086/694039. Dorsch, J., K.Goddu, K.Nave, T.Vierkant, M.Coeckelbergh, P.Gürtler, P.Urban, F.Spang & M.Moll (2025). Against AI Welfare: Care Practices Should Prioritize Living Beings Over AI. AI Magazine46: e70016. Douglas, H. (2000). Inductive Risk and Values in Science. Ph...

    2025 doi arXiv

  12. [165]

    Malfacini, K

    10.1007/s11098-012-9930-7. Malfacini, K. (2025). The impacts of companion AI on human relationships: risks, benefits, and design considerations. AI & Soc. https://doi.org/10.1007/s00146-025-02318-6 Mallory, F. (2023). Fictionalism about Chatbots. Ergo an Open Access Journal of...

    2025 doi

  13. [176]

    Reinecke, M

    10.1007/s11098-018-1038-2. Reinecke, M. G., Ting, F., Savulescu, J., & Singh, I. (2025). The Double-Edged Sword of Anthropomorphism in LLMs. Proceedings, 114(1),

    2025 doi

  14. [180]

    Steenkamp, Jan-Benedict & Jong, Martijn & Baumgartner, Hans

    1-31. Steenkamp, Jan-Benedict & Jong, Martijn & Baumgartner, Hans. (2010). Socially Desirable Response Tendencies in Survey Research. Journal of Marketing Research - J MARKET RES-CHICAGO

    2010

  15. [1994]

    god-like

    and may be monothematic (concerning a single theme) or polythematic (concerning various themes, e.g., feeling persecuted, on a divine mission, etc.) (Coltheart, 2013). Similarly, when consciousness attributions to chatbots express non-bizarre delusions, the related interferenc...

    2013

  16. [2013]

    https://www.theguardian.com/world/2013/apr/02/americans-obama-anti-christ-conspiracy-theories. Heyes C. (2018). Empathy is not in our genes. Neuroscience and biobehavioral reviews, 95, 499–507. 22 Hudson R. (2021). Should We Strive to Make Science Bias-Free? A Philosophical As...

    2013

  17. [2015]

    This might happen, for instance, by selectively attending to, discounting, or rationalizing the counterevidence away

    or motivated reasoning (Kunda, 1990), serving truth-unrelated goals, including psychological comfort, companionship, social belonging (e.g., to a Replika community), or economic and reputational incentives (e.g., to promote products, or attract investment), may lead chatbot us...

    1990

  18. [2017]

    illusion of objectivity

    and the belief remains counterevidence responsive, in such biased belief cases, the difference doesn’t lie in the threshold for belief but in how its maintenance is governed. Unbiased belief is constrained by evidence whereas biased belief is stabilized against it, indicating ...

    1990

  19. [2021]

    Association for Computing Machinery, New York, NY, USA, 610–623

    On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (FAccT '21). Association for Computing Machinery, New York, NY, USA, 610–623. Bennett, J. (1990). Why Is Belief Involuntar...

    2021

  20. [2023]

    cares.”4 Since caring and empathy are widely understood as being able to “feel what someone else is feeling

    that “cares.”4 Since caring and empathy are widely understood as being able to “feel what someone else is feeling”, positive or negative (Heyes, 2018, p. 499), such claims indirectly attribute sentience to the model. Yet, on a subpage, the company owner(s) also notes that “Rep...

    2018

  21. [2025]

    first-mover advantage

    and propose policies reflecting this concern (e.g., Anthropic) may secure a “first-mover advantage” (Lieberman, 2016), presenting themselves as ethically prepared, which can appeal to users, policymakers, and investors (Edwards, 2026). Strategic pretence resembles Frankfurtian...

    2016

  22. [2026]

    epistemic innocence

    or that the evidence remains inconclusive (McClelland, 2025), are people epistemically blameworthy when they nonetheless attribute consciousness to such systems? Should we try to correct them for their attributions? To make progress on these questions, my goal here is twofold....

    2025

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.