Deception

Human deception involves intentionally causing false beliefs. Since it is contested whether large language models (LLMs) and artificial intelligence (AI) in general have beliefs and intentions, deception by such systems is typically defined in terms of observable behavior. However, this paper argues that such behavioral definitions fall short of the strategic nature of deception, among other things, because they do not consistently distinguish deception from mere error. Moreover, behavioral definitions of AI deception are not optimal as methodological guidance for mechanistic (...) approaches to AI deception because they do not specify candidates for internal markers of deception. It is important to take the internal mechanisms of LLMs and AI systems into account when addressing deception, since skilled forms of deception may be behaviorally indistinguishable from honesty. This paper formulates these challenges to behavioral definitions and addresses them by proposing a novel definition of strategic AI deception in terms of generalized propositional attitudes. Specifically, it provides generalizations of the attitudes of belief and desire that can be ascribed to current and future AI systems less controversially than their human counterparts, yet are expressive enough to figure in a definition of strategic deception. This definition has several advantages over its behavioral alternatives as it 1) better satisfies pre-theoretic intuitions such as the error condition 2) fits better with applied researchers’ usage of the term “strategic deception”, 3) is relevant to large-scale risks of AI deception, and 4) is methodologically instructive for mechanistic approaches to deception. (shrink)

According to an increasingly influential view, the complex, human-like behavior of large language models (LLMs) should be explained as roleplay or simulation of characters. Such notions aim at enabling intentional explanations without ascribing mental states to LLMs. However, there are a multitude of roleplay views that are applied inconsistently in empirical explanations of LLM behavior. Moreover, it is unclear what exactly different roleplay views predict and, consequently, how they can be empirically distinguished from each other and from explanations that do (...) ascribe mental states to LLMs. In this paper, we provide a taxonomy of roleplay views structured through three dimensions: Interpretation (how can LLM behavior be best explained according to the view), Mechanism (what commitments does the view have about internal mechanisms of roleplay), and Generality (does the view aim to explain all LLM behavior or specific kinds). We then analyze, with regard to the case of LLM deception, why it is practically relevant whether LLMs are mere roleplayers or in possession of genuine mental states. However, we identify two challenges for empirically distinguishing these two: Similarity (Roleplay is inherently difficult to distinguish from intentional behavior) and Ambiguity (it is often ambiguous what roleplay views predict). Finally, we propose three strategies for overcoming these challenges, making progress towards transforming roleplay from loose metaphor to a testable theory. (shrink)

We all have been deceived, whether by politicians, pundits or peers. Likely more often than we would wish to admit. But what is it to be deceived? As we argue, victims of deception settle on a false answer to a question. This settling proposal not only makes good on the intuitive idea that sceptical agents are harder to deceive than their non-sceptical peers, but also finds support in the infelicity of reporting an agent as both inquiring and deceived. Want to (...) avoid being duped? Our proposal provides an answer: inquire further, don't settle. (shrink)

The central claim of this paper is that chatbots are, in an important sense, deceptive technologies; this deception, however, is not always morally problematic. It becomes so when it is about the relationship the user has with the bot. Since our relationship with chatbots is built solely on dialogue, the ethics of dialogue can offer a privileged lens for discerning harmless and harmful deception. It emerges that the ethical issue with chatbot interactions does not lie in users being deceived about (...) the system’s nature, i.e., thinking it is human, but rather in being misled about the quality of the dialogue they are engaging in. In particular, it is morally bad when chatbots deceive the user about the mutual openness of their interaction, as such systems are inherently incapable of reciprocating. Not all chatbots engage in harmful deception: some remain within the boundaries of what is known as banal deception. Broadly speaking, it is possible to distinguish between two categories of bots: instrumental chatbots, which are task-driven and engage only in banal deception, and simulated-behaviour chatbots, which may engage in harmful deception. In this latter category, AI companions are always morally problematic as they intentionally deceive users about the mutuality of the relationship. Based on this diagnosis, some normative prescriptions are advanced to regulate AI companions and ensure they foster only banal deception. The paper contributes both conceptual clarity to the ethics of AI deception and practical guidance for the responsible design and governance of chatbot technologies. (shrink)

There is increasing interest in the use of artificial intelligence for companionship. Most resistance to this proposal appeals to the risk of deception: the idea that, because they lack the relevant mental states, artificial companions cannot authentically befriend their users. In this paper, I argue that it would be morally impermissible to befriend artificial companions even if there was no deception involved in doing so. This is because, without deception, such befriending would become exploitative: it would require users to take (...) unjust advantage of certain vulnerabilities required for authentic friendship. I make the premises of this argument explicit and defend each in turn, both positively and from various objections. The resulting view is one on which—barring some interesting alternatives raised by those objections—non-deceptive artificial companionship is analogous to knowing participation in an arranged marriage. (shrink)

Virtual reality headsets present us with two images, one directly before each of our eyes. One might therefore suppose that virtual reality offers experiences with the phenomenology of many other depictive images – an experience of seeing-in. This paper argues that the phenomenology of virtual reality does not involve seeing-in due to our lacking the required awareness of virtual reality images’ configuration. Instead, virtual reality is intended to generate an experience resembling that of everyday perception, artificially recreating the phenomenology of (...) various experiences. Interestingly, however, the present technical limitations of virtual reality frequently also generate an underrecognized form of twofold experience, where we are perceptually aware of both the content of a virtual reality image and the way that content is represented, yet still without our seeing-in. But while this twofold experience is often unintended, it can also be deliberately aimed at to help guard against some of virtual reality’s potential harms. (shrink)

Gaslighters make their targets feel defective for possessing mental states to which they are entitled. This kind of deceptive manipulation is universally condemned as wrongful and destructive by philosophers and psychologists, as it destroys its target’s epistemic agency and psychological well-being and can be epistemically unjust. But these issues apply to gaslighting that is downward-facing—where the powerful gaslight the vulnerable—and gaslighting is not always like this. When the traditional power dynamic is reversed and a vulnerable agent gaslights her powerful oppressor (...) to defend herself from being wronged, I call it gaslighting-up and argue that such behavior is morally permissible and epistemically valuable. It enables abuse victims to avoid harm and to resist the very epistemic domination that is caused by abusive gaslighting, enables marginalized agents to safely protest bigotry and avoid the kind of testimonial injustice that they would face from traditional epistemic gaslighting, and even epistemically benefits its dogmatic targets by giving them a healthy dose of epistemic humility. So, rather than condemning gaslighting as intrinsically, universally bad, we should take a nuanced view and consider it in the context of real-world power dynamics and the role it plays in reinforcing and destabilizing these. (shrink)

Our received theories of self-deception are problematic. The traditional view, according to which self-deceivers intend to deceive themselves, generates paradoxes: you cannot deceive yourself intentionally because you know your own plans and intentions. Non-traditional views argue that self-deceivers act (sub-)intentionally but deceive themselves unintentionally and unknowingly. Some non-traditionalists even say that self-deception involves a mere error (of self-knowledge). The non-traditional approach does not generate paradoxes, but it entails that people can deceive themselves by accident or by mistake, which is rather (...) controversial. I argue that a functional analysis of human interpersonal deception and self-deception solves both problems and a few more. According to this analysis, my behavior is deceptive iff its function is to mislead; I may but need not intend to mislead. In self-deception, then, the self engages in some deceptive behavior and this behavior misleads the self. Thus, while it may but need not be intended, self-deception is never an accident or a mistake. (shrink)

Scientifically, stats are born of trying to equate what is not equal through an 'equation', obviously.

Quite early in the Covid-19 pandemic, a recommendation was issued in Germany to address potential scarcity scenarios in hospital intensive care units. At its core, the recommendation from Germany’s medical professional societies stated that, in the event of overcrowded ICUs, physicians should base the selection of patients who could still be admitted on the likelihood of success for each individual in need (DIVI 2021). The purpose of focusing on the likelihood of success is to use the available resources to help (...) the greatest possible number of people in need effectively. However, legal scholars have repeatedly argued that a lottery-based solution would be far more suitable for such tragic rescue conflicts. This perspective was also adopted by some disability advocates, who fear the risk of discrimination against people with disabilities in scarcity scenarios, as their chances of success are often presumed to be systematically lower than those of people without disabilities. From an ethical standpoint, this raises the legitimate question of the conditions under which a decision-making criterion implies indirect discrimination against a particular group of people. ----- Schon recht früh in der Covid-19-Pandemie wurde in Deutschland eine Empfehlung für mögliche Knappheitsszenarien auf den Intensivstationen der Krankenkenhäuser herausgegeben. Im Kern besagt die Empfehlung der deutschen medizinischen Fachgesellschaften, die Ärzteschaft solle im Falle überfüllter Intensivstationen die Auswahl derjenigen, die in einem solchen Fall nochmal aufgenommen werden können, an der Erfolgsaussicht der einzelnen Bedürftigen festmachen. Der Zweck einer Orientierung an der Erfolgsaussicht besteht darin, mit den vorhandenen Ressourcen einer möglichst großen Anzahl an Bedürftigen erfolgreich helfen zu können. Insbesondere von Juristen wurde jedoch auch immer wieder argumentiert, eine Lotterie-Lösung sei sehr viel geeigneter für derartige tragische Rettungskonflikte. Eine solche Auffassung wurde auch von einigen Behindertenvertreter:innen übernommen, die eine Gefahr der Benachteiligung von Menschen mit Behinderung in Knappheitsszenarien befürchten, denn bei ihnen dürften die Erfolgsaussichten regelhaft schlechter sein als bei Menschen ohne Behinderung. In ethischer Hinsicht wurde hier also die berechtigte Frage aufgeworfen, unter welchen Bedingungen ein Entscheidungskriterium eine indirekte Diskriminierung einer bestimmten Personengruppe impliziert. (shrink)

The police are routinely recognized for displaying heroic virtues associated with combat. I take a contrarian position in this paper. Part I begins with the claim that if bravery is to be prioritized in policing, then bravery should be part of the police’s routine roles and responsibilities. However, bravery is not central to what the police do every day, and, therefore, shouldn’t be prioritized (in recruiting, training, and so on). Conversely, Part II claims that if the virtue of honesty is (...) centrally connected to the police role, then it should be prioritized in policing. Because honesty is deeply connected to the police’s routine roles and responsibilities, it should be prioritized. By turning away from misconceptions about the police, the hope is that we will be in a better position to see both the practical and the moral importance of virtue in policing. (shrink)

How is it that a speaker _S_ can at once make it obvious to an audience _A_ that she intends to communicate some proposition _p_, and yet at the same time retain plausible deniability with respect to this intention? The answer is that _S_ can bring it about that _A_ has a high justified credence that ‘_S_ intended _p_’ without putting _A_ in a position to know that ‘_S_ intended _p_’. In order to achieve this _S_ has to exploit a (...) sense in which communication can be lottery-like. After defending this view of deniability I argue that it compares favorably to a rival account recently developed by Dinges and Zakkou (Mind, https://doi.org/10.1093/mind/fzac056, 2023 ). (shrink)

The use of deception and dishonesty is widely accepted as a fact of life in policing. This paper thus defends a counterintuitive claim: Good faith is a normative foundation for the police as a political institution. Good faith is a core value of contracts, and policing is contractual in nature both broadly (as a matter of social contract theory) and narrowly (in regard to concrete encounters between law enforcement officers and the public). Given the centrality of good faith to policing, (...) dishonesty and deception on par with fraud are justified only as a narrowly circumscribed investigative tool that is constrained by institutional commitments to the fair distribution of security and the rule of law. The practical upshot is the preclusion of most dishonest and deceptive police tactics on par with fraud, leading to an institution that is less proactive and more reactive. (shrink)

Regarding the appropriateness of deception in clinical practice, two (apparently conflicting) claims are often emphasised. First, that ‘clinicians should not deceive their patients.’ Second, that deception is sometimes ‘in a patient’s best interest.’ Recently, Hardman has worked towards resolving this conflict by exploring ways in which deceptive and non-deceptive practices extend beyond consideration of patients’ beliefs. In short, some practices only seem deceptive because of the (common) assumption that non-deceptive care is solely aimed at fostering true beliefs. Non-deceptive care, however, (...) relates to patients’ non-doxastic attitudes in important ways as well. As such, Hardman suggests that by focusing on belief alone, we sometimes misidentify non-deceptive care as ‘deceptive’. Further, once we consider patients’ beliefs and non-doxastic attitudes, identifying cases of deception becomes more difficult than it may seem. In this essay, I argue that Hardman’s reasoning contains at least three serious flaws. First, his account of deception is underdeveloped, as it does not state whether deception must be intentional. The problem is that if intention is not required, absurd results follow. Alternatively, if intention is required, then identifying cases of deception will be much easier (in principle) than Hardman suggests. Second, Hardman mischaracterises the ‘inverse’ of deceptive care. Doing so leads to the mistaken conclusion that common conceptions of non-deceptive care are unjustifiably narrow. Third, Hardman fails to adequately separate questions about deception from questions about normativity. By addressing these issues, however, we can preserve some of Hardman’s most important insights, although in a much simpler, more principled way. (shrink)

This book proposes a novel theory of truth and falsity. It argues that truth is a form of reference and falsity is a form of reference failure. -/- Most of the philosophical literature on truth concentrates on certain ontological and epistemic problems. This book focuses instead on language. By utilizing the Fregean idea that sentences are singular referring expressions, the author develops novel connections between the philosophical study of truth and falsity and the huge literature in in the philosophy of (...) language on the notion of reference. The first part of the book constructs the author’s theory and argues for it in length. Part II addresses the ways in which the theory relates to, and is different from, some of the basic theories of truth. Part III takes up how to account for the truth of sentences with logical operators and quantifiers. Finally, Part IV discusses the applications and implications of the theory for longstanding problems in philosophy of language, metaphysics, and epistemology. -/- A Referential Theory of Truth and Falsity will appeal to researchers and advanced students working in philosophy of language, epistemology, metaphysics, and linguistics. -/- . (shrink)

In this paper I point to the degraded value of hope in the contemporary digital world and discuss the dangers of this transformation. Hope constitutes one of the basic drives that sustains vitality in the human species. Especially contemporary philosophers such as Hume, Kant, and Bloch, among others, consider hope from a philosophical perspective and despite their differences, they agree on the overall importance of hope as one of the fundamental motivations of humans towards a future life that makes striving (...) possible. However, in the contemporary world, starting with the film industry as Adorno and Horkheimer stated, and later advanced by social media practices, hope becomes solely a hope of fame. This desire for fame is such that, as Baudrillard anticipated before the widespread use of the internet and the emergence of social media, a panoptical system is no longer required as people share their private life publicly, 'hoping' that they gain more followers, likes and attention. People exhibit 'glamorous', 'beautiful', and 'delicious' moments that constrain hope to one dimension and disconnect it from reality. From these points, I argue that what the contemporary world presents is a passive, negative hope that marks the loss of hope which should be productive and transformative. (shrink)

Drawing from narratology and design studies, this article makes use of the notions of the ‘implied designer’ and ‘ludic unreliability’ to understand deceptive game design as a specific sub-set of transgressive game design. More specifically, in this text we present deceptive game design as the deliberate attempt to misguide players’ inferences about the designers’ intentions. Furthermore, we argue that deceptive design should not merely be taken as a set of design choices aimed at misleading players in their efforts to understand (...) the game, but also as decisions devised to give rise to experiential and emotional effects that are in the interest of players. Finally, we propose to introduce a distinction between two varieties of deceptive design approaches based on whether they operate in an overt or a covert fashion in relation to player experience. Our analysis casts light on expressive possibilities that are not customarily part of the dominant paradigm of user-centered design, and can inform game designers in their pursuit of wider and more nuanced creative aspirations. (shrink)

Michael Goldhaber, who popularized the term the attention economy, said of the US Capitol insurrection: “It felt like an expression of a world in which everyone is desperately seeking their own audience and fracturing reality in the process. I only see that accelerating.” If we don’t do something about this, American democracy may not survive. For when there is no longer any common ground of evidence and reason, history shows that misinformation will eventually overwhelm public discourse and authoritarianism can take (...) over. That is precisely what dictators across the world gleefully anticipate will happen in the US. The Biden administration could combat this by establishing a digital fairness doctrine for the twenty-first century. This would not mean setting up a government agency as the arbiter of truth. The purpose of such a doctrine is simply to preserve the possibility of a functioning national discourse on critical issues of public interest. It would not run afoul of First Amendment free speech guarantees, given that it would do nothing to block speech itself. In fact, it would expand and enrich national conversations, by preventing echo chambers prone to systematic bias. (shrink)

The feeling of clarity can be dangerously seductive. It is the feeling associated with understanding things. And we use that feeling, in the rough-and-tumble of daily life, as a signal that we have investigated a matter sufficiently. The sense of clarity functions as a thought-terminating heuristic. In that case, our use of clarity creates significant cognitive vulnerability, which hostile forces can try to exploit. If an epistemic manipulator can imbue a belief system with an exaggerated sense of clarity, then they (...) can induce us to terminate our inquiries too early — before we spot the flaws in the system. How might the sense of clarity be faked? Let’s first consider the object of imitation: genuine understanding. Genuine understanding grants cognitive facility. When we understand something, we categorize its aspects more easily; we see more connections between its disparate elements; we can generate new explanations; and we can communicate our understanding. In order to encourage us to accept a system of thought, then, an epistemic manipulator will want the system to provide its users with an exaggerated sensation of cognitive facility. The system should provide its users with the feeling that they can easily and powerfully create categorizations, generate explanations, and communicate their understanding. And manipulators have a significant advantage in imbuing their systems with a pleasurable sense of clarity, since they are freed from the burdens of accuracy and reliability. I offer two case studies of seductively clear systems: conspiracy theories; and the standardized, quantified value systems of bureaucracies. (shrink)

Shills are people who endorse products and companies for pay, while pretending that their endorsements are ingenuous. Here we argue that there is something objectionable about shilling that is not reducible to its bad consequences, the lack of epistemic conscientiousness it often relies upon, or to the shill’s insincerity. Indeed, we take it as a premise of our inquiry that shilling can sometimes be sincere, and that its wrongfulness is not mitigated by the shill’s sincerity, in cases where the shill (...) is sincere. Our proposal is that the shill’s defining characteristic is their knowingly engaging in a kind of speech that obscures a certain aspect of its social status – most commonly, by pretending to speak on their own personal behalf, while in fact speaking as an employee – and that this sort of behaviour is objectionable irrespective of any other features of the shill’s conduct. This sort of obfuscation undermines a socially beneficial communicative custom, in which we conscientiously mark the distinction between personal speech and speech-for-hire. (shrink)

This paper addresses two questions about audience misunderstandings of contemporary art. First, what is the institution’s responsibility to prevent predictable misunderstandings about the nature of a contemporary artwork, and how should this responsibility be balanced against other considerations? Second, can an institution ever be justified in intentionally mounting an inauthentic display of an artwork, given that such displays are likely to mislead? I will argue that while the institution has a defeasible responsibility to mount authentic displays, this is not always (...) sufficient to avoid misunderstanding; the institution will sometimes need to supply auxiliary information. And even where competing considerations require mounting an inauthentic display, thoughtful museum practice can promote the audience’s ability to grasp the work. The argument will be developed with consideration of artworks by El Anatsui, Lygia Clark, and Glenn Ligon. (shrink)

This paper introduces and argues for the hypothesis that judgments of testimonial worth are central to our practice of normatively appraising speech. It is argued that judgments of testimonial worth are central both to the judgement that an agent has lied, and to the acceptance of testimony. The hypothesis that, in lying, an agent necessarily displays poor testimonial worth, is shown to resolve a new puzzle about lying, and the recalcitrant problem raised by the existence of bald faced lies, and (...) selfless assertions. It is then shown that the notion of testimonial worth allows us to capture the distinction between taking a speaker at their word, and treating them as a mere indicator of the truth in a way other theories fail to do. (shrink)

This book on the topic of ethics and poetry consists of contributions from different continents on the subject of applied ethics related to poetry. It should gather a favourable reception from philosophers, ethicists, theologians and anthropologists from Africa, Asia, Europe and Latin America and allows for a comparison of the healing power of words from various religious, spiritual and philosophical traditions. The first part of this book presents original poems that express ethical emotions and aphorism related to a philosophical questioning (...) of the grounding of our values for life. The poems are written by twelve authors coming from four continents, for whom poetic emotions are sources of artistic inspiration and that can be used for conflict resolution. In the second part, which features short essays, nine authors tackle how poems, symbolic representations, metaphorical narratives and lies impact the space of possibilities, in which we are moved to action, knowledge formation, and how we imagine the world together. (shrink)

Games have a complex, and seemingly paradoxical structure: they are both competitive and cooperative, and the competitive element is required for the cooperative element to work out. They are mechanisms for transforming competition into cooperation. Several contemporary philosophers of sport have located the primary mechanism of conversion in the mental attitudes of the players. I argue that these views cannot capture the phenomenological complexity of game-play, nor the difficulty and moral complexity of achieving cooperation through game-play. In this paper, I (...) present a different account of the relationship between competition and cooperation. My view is a distributed view of the conversion: success depends on a large number of features. First, the players must achieve the right motivational state: playing for the sake of the struggle, rather than to win. Second, successful transformation depends on a large number of extra-mental features, including good game design, and social and institutional features. (shrink)

In this essay I argue that Machiavelli does not hold that all deception is permissible in war. While Machiavelli claims that "deceit... in the conduct of war is laudable and honorable," he insists that such deceit, or ruses of war, is not to be confounded with perfidy. Any Lee's U.S. Civil War film, "Ride With the Devil," illustrates this difference. The film also illustrates the difference between lying as part of romance, which is permitted, and lying at the moment of (...) truth in a relationship, when admitting one's feelings, which is not. (shrink)

"Crooked people deceive themselves in order to deceive others; in this way the world comes to ruin." This quote from a medieval Confucianist expresses the ethical danger of self-deception. My paper examines the psychological proclivity for self-deception and argues that it lies behind much social and interpersonal injustice. I review Hitler's Mein Kampf, as a premiere example of such cognitive duplicity, and Socratic dialectic, as an example of the cognitive hygiene necessary to combat it. I conclude that a robust educational (...) program of Socratic-style critical thinking is crucial to the furtherance of a just society. (shrink)

In the Republic, Socrates argues that morality (justice) is valuable both for itself and for what comes from it. In contemporary moral theory, this view is not widely accepted. However, contemporary empirical research in psychology reveals that what we experience is also what we come to expect. It follows from this that if we act in an immoral fashion, we will expect the same from others. The more often we act immorally, the more suspicion will be ingrained within us. Suspicion (...) quickly pervades our minds, corrodes our relationships, and increases our levels of stress, thereby dramatically decreasing our quality of life. Moral action, on the other hand, has the ability to strengthen our faith in others and create strong human relationships. Therefore,morality should be classified as having the highest type of value, just as Socrates contends.Moral behavior allows for healthy relations between persons (intrinsic value) and the wellbeing of the individual (extrinsic value). (shrink)

Every day, television news operations have available dozens of video news releases (VNRs), public relations handout videos designed to mimic news formats. Electronic tracking indicates some of these VNRs are used. Critics typically assail VNRs on ethical grounds, that VNRs deceive audience members into thinking they are watching news gathered by reporters, rather than a promotional pitch. Using a snowball technique, the researchers presented survey respondents with authentic-looking local television news stories; 157 respondents evaluated three stories (out of nine). Some (...) stories used station-generated footage, some network, and some VNRs. Respondents were asked their best estimation of the source. The data indicated a real likelihood VNR deception is occurring. Two of the three VNRs deceived at least half the respondents. Respondents were incorrect an average of 46% of the time when attempting to identifying VNRs, compared to 29% incorrectly identifying network video and 20% incorrectly identifying locally shot video. (shrink)

The hidden brain has its finger on the scale when we make all of our most complex and important decisions – it decides who we fall in love with, whether we should convict someone of murder, or which way to run when someone yells “fire ...

The aim of this paper is two-fold. We argue against the received conception of epistemic malevolence and give a broader characterisation that, we argue, captures its real scope. We tackle the current notion of epistemic malevolence (EM) on three fronts. We claim that this notion fails to capture cases of EM that are (i) not knowledge directed (but that affect the agent's epistemic character), (ii) not robustly volitional (that do not involve an active opposition to the epistemic good by the (...) malevolent agent) and that (iii) are not merely transactional (as opposed to structural). As a result, we claim that the received notion of EM fails to recognise (arguably) clear instances of EM because it is too narrow. Ultimately, we argue that such narrowness gives the false impression that, unlike its moral counterpart, EM is uncommon. After arguing for this, we offer a wider characterisation of EM and show the role this can play in generating solutions to the problem the phenomena poses. (shrink)

Sincerity is essential to communication: without a norm of sincerity, we could hardly trust what other people tell us. But what does it take to be sincere, exactly? And why is sincerity so important? Sincerity and Insincerity offers a comprehensive review of existing philosophical work on the nature of sincerity and its epistemic value. It puts forward a novel, fine-grained account of what sincerity and insincerity are, and dives into the grey area between the two, identifying various ways in which (...) speakers can be partially sincere. Integrating ideas from different philosophical subfields and traditions, it offers an updated perspective on what makes sincerity epistemically valuable, giving serious consideration to the idea that sincerity is the norm of assertion. Overall, this Element provides a novel, informed perspective on what sincerity is, how it works, and why it matters. (shrink)

Can Large Language Models (LLMs) perform speech acts? I show how a norm-based argument against LLM-assertion can be generalized so as to support a negative answer to this question. I then investigate how the resulting no-speech-act-view interacts with several other debates about how to conceptualize the linguistic outputs of LLMs. I argue that it conflicts with the popular conception of LLMs as bullshitters and that it helps decide between two competing externalist accounts of how artificially generated expressions refer. I close (...) by discussing a challenge to the no-speech-act-view and by suggesting a novel strategy for addressing it. A general lesson from this discussion is that our understanding of LLMs’ linguistic outputs will profit from complementing extant meta-semantic theorizing with more detailed meta-pragmatic theorizing about discourse between humans and LLMs. (shrink)

This paper defends the thesis that LLM hallucinations are best explained as a truth representation problem: Current models lack an internal representation of propositions as truth-bearers, so truth and falsity cannot constrain generation in the way factual discourse requires. It begins by surveying leading explanations—computational limits on self-verification, deficiencies in training data as truth sources, and architectural factors—and argues that they converge on the same underlying representational deficit. Next, it reconstructs the philosophical background of current LLM design, showing how optimization (...) for fluent continuation aligns with coherence-style evaluation and with broadly structuralist, relational semantics, before turning to David Chalmers’s recent attempt to secure propositional interpretability by drawing on Davidson/Lewis-style radical interpretation and by locating propositional content in “middle-layer” structures; it argues that this approach downplays the ubiquity of hallucination and inherits instability from post-training edits. Finally, the paper offers a positive proposal: Atomic propositions should be represented in the basic vector layer, reviving a logical atomist program as a principled route to reducing hallucination. (shrink)

In Mizrahi (2025a), I sketch two arguments against the claim that LLMs should be granted the status of epistemic authorities worthy of epistemic respect. Hauswald (2026) objects to the second premise of each of these arguments and asks, “Why shouldn’t there be reliable bullshit machines?” In this paper, I respond to his objections and answer his question. Basically, if bullshitting is not a reliable, truth-conducive process, then there should not, indeed could not, be reliable bullshit machines.

There is a certain obscurantist craft by which a speaker’s words, deep or profound as they may seem, are so unclear that they frustrate the interpreter’s hermeneutical efforts. In the end, the interpreter may come to doubt her hermeneutical competencies while inflating her perception of the obscurantist’s epistemic status. The interpreter epistemically subordinates herself, in this way, to the obscurantist. In this paper, I elucidate the nature of this ‘subordinative obscurantism’ and examine its epistemically manipulative applications. I also discuss how (...) subordinative obscurantist manipulation can be used in conjunction with another epistemic manipulation strategy rooted in the use of clear-seeming speech, thereby setting especially pernicious cognitive traps for victims. Finally, I discuss the epistemic status of the beliefs formed by victims of subordinative obscurantist manipulation. (shrink)

We develop an account of bullshit questions that draws on the literature on bullshit assertions. We distinguish bullshit questions from other sorts of anomalous questions. According to our account, bullshit questions are characterized chiefly by the indifference of the speaker to the truth of any answer she might receive. Instead, the bullshit questioner is up to something else, typically a non-interrogative illocutionary act such as introducing a presupposition, insinuating a derogatory sentiment, implying a proposition, making an accusation, or flirting. If (...) this is right, it naturally raises the normative question of whether and how bullshit questions are wrong and whether and how bullshit questioners are blameworthy and vicious. In the final section, we address these questions, arguing that bullshit questions are pro tanto wrong because they tend to thwart inquiry, manifest the vice of epistemic insouciance (which is a disregard for truth or inquiry), express disrespect for the epistemic agency of the interlocutor, introduce epistemic malaise, and lead to the opening of dangerous inquiries by gullible audiences. We then consider some cases in which the pro tanto wrongness of bullshit questioning is arguably overridden by competing reasons. (shrink)

Political leaders may change their mind about a policy, or even a significant moral issue. While genuinely changing one’s mind is not hypocritical, there are reasons to think that leaders who claim such a change are merely hypocritically pandering for political advantage. Indeed, some social science studies allegedly confirm that constituents will judge political leaders who change positions as hypocritical. Yet these studies are missing crucial details that we normally use to distinguish genuine mind changers from hollow hypocrites. These details (...) suggest ways in which political leaders can communicate that their mind change is genuine without necessarily being judged hypocritical. (shrink)

This paper argues that questions can be bullshit. First it explores some shallowly interrogative ways in which that can happen. Then it shows how questions can also be bullshit in a way that’s more deeply interrogative.

It has become almost a cliché to say that we live in a post-truth world; that people of all trades speak with an indifference to truth. Speaking with an indifference to how things really are is famously regarded by Harry Frankfurt as the essence of bullshit. This paper aims to contribute to the philosophical and theoretical pragmatics discussion of bullshit. The aim of the paper is to offer a new theoretical analysis of what bullshit is, one that is more encompassing (...) than Frankfurt’s original characterization. I part ways with Frankfurt in two points. Firstly, I propose that we should not analyze bullshit in intentional terms (i.e. as indifference). Secondly, I propose that we should not analyze it in relation to truth. Roughly put, I propose that bullshit is best characterized as speaking with carelessness toward the evidence for one’s conversational contribution. I bring forward, in the third section, a battery of examples that motivate this characterization. Furthermore, I argue that we can analyze speaking with carelessness toward the evidence in Gricean terms as a violation of the second Quality maxim. I argue that the Quality supermaxim, together with its subordinate maxims, demand that the speaker is truthful (contributes only what she believes to be true) and reliable (has adequate evidence for her contribution). The bullshitter’s main fault lies in being an unreliable interlocutor. I further argue that we should interpret what counts as adequate evidence, as stipulated by the second Quality Maxim, in contextualist terms: the subject matter and implicit epistemic standards determine how much evidence one needs in order to have adequate evidence. I contrast this proposed reading with a subjectivist interpretation of what counts as having adequate evidence and show that they give different predictions. Finally, working with a classic distinction, I argue that we should not understand bullshit as a form of deception but rather as a form of misleading speech. (shrink)

‘Fake news’ has become an increasingly common refrain in public discourse, though the term itself has several uses, at least one of which constitutes Frankfurtian bullshit. After examining what sorts of fake news appeals do and do not count as bullshit, I discuss strategies for overcoming our openness to such bullshit. I do so by drawing a parallel between openness to bullshit and naïve skepticism—one’s willingness to reject the concept of truth on unsupported or ill-considered grounds—and suggest that this parallel (...) indicates three principles for how we ought to combat our openness to fake news and other bullshit. First, the root causes of bullshit openness are not monolithic; we should adopt anti-bullshit strategies in recognition of this fact. Second, our efforts to overcome bullshit openness should be collaborative efforts to create an environment that allows for sustained interrogation of our bullshit openness, rather than a confrontational provision of contrary evidence, despite the fact that such strategies are more time-intensive. Third, social media is unlikely to be a fertile ground on which we will make meaningful progress in the fight against bullshit because of the inherent nature of social media platforms as spaces for short, declarative, confrontational claims. (shrink)

This thesis addresses philosophical problems concerning improper assertions. The first part considers the issue of defining lying: here, against a standard view, I argue that a lie need not intend to deceive the hearer. I define lying as an insincere assertion, and then resort to speech act theory to develop a detailed account of what an assertion is, and what can make it insincere. Even a sincere assertion, however, can be improper (e.g., it can be false, or unwarranted): in the (...) second part of the thesis, I consider these kinds of impropriety. An influential hypothesis maintains that proper assertions must meet a precise epistemic standard, and several philosophers have tried to identify this standard. After reviewing some difficulties for this approach, I provide an innovative solution to some known puzzles concerning the permissibility of false assertions. In my view, assertions purport to aim at truth, but they are not subject to a norm that requires speakers to assert a proposition only if it is true. (shrink)

This paper refines a number of theoretical distinctions relevant to deceptive play, in particular the difference between merely misleading actions and types of simulation commonly considered beyond the pale, such as diving. To do so, I rely on work in the philosophy of language about conversational convention and implicature, the distinction between lying and misleading, and their relation to concepts of seduction and bullshit. The paper works through a number of possible solutions to the question of what is wrong with (...) simulation and its difference from strategic fouling, including the argument that games and their rules operate like contracts. I conclude that the wrongness lies in the injustice of unfair advantage gained through actions that silence opposition by resort to unanswerable play. (shrink)

Sincerity is essential to communication: without a norm of sincerity, we could hardly trust what other people tell us. But what does it take to be sincere, exactly? And why is sincerity so important? Sincerity and Insincerity offers a comprehensive review of existing philosophical work on the nature of sincerity and its epistemic value. It puts forward a novel, fine-grained account of what sincerity and insincerity are, and dives into the grey area between the two, identifying various ways in which (...) speakers can be partially sincere. Integrating ideas from different philosophical subfields and traditions, it offers an updated perspective on what makes sincerity epistemically valuable, giving serious consideration to the idea that sincerity is the norm of assertion. Overall, this Element provides a novel, informed perspective on what sincerity is, how it works, and why it matters. (shrink)

A great deal of attention has been directed at the question of what exactly is required for an utterance to count as a lie. At the center of recent discussion stand bald-faced lies, which have proven to be remarkably resistant to philosophical analysis. This chapter focuses on a related, yet curiously under-explored, set of cases: lies that we construct together, as friends, families, colleagues, and communities. This sort of lie exhibits a degree of moral and linguistic complexity not found in (...) more standard examples of lying. That moral complexity will ultimately put pressure on the enduring thesis that the distinctive wrong of lying is that it threatens to undermine the potential for communication. The linguistic complexity, in contrast, will stand as a challenge to standard theories of conversational dynamics. (shrink)

According to a recent account of disinformation, disinformation is content that “generates ignorance” (Simion 2024a; 2024b). The view improves upon previous accounts that focused upon the potential for disinformation to induce false belief, overlooking its role in generating ignorance by inducing doubt. While this proposal gives us a broader understanding of what disinformation can be, it retains the idea that disinformation functions as evidence that incrementally confirms falsehoods. Thus, this approach implies (in line with previous views) that when disinformation is (...) used intentionally, the disinformant hopes to capture the attention of agents and induce these agents to process this content as evidence to change an agent's belief state in epistemically undesirable ways. Certainly, this is true of some disinformation tactics, but it won't give us an adequate definition. In some successful real-world disinformation campaigns, disinformants add to the available content to deter the audience from attending to and processing evidence. In some cases, evidence can incrementally confirm a falsehood while providing knowledge of its negation. I propose an alternative pragmatic view that explains this. On this view, disinformation can come in the form of misleading evidence, but it can also come in the form of content that makes it evident that the evidential resources the agent needs to advance inquiry have been corrupted. This content functions as disinformation because it rationalises decisions that can protect a disinformant’s interests instead of promoting the inquirer’s interests. Instead of thinking of disinformation in epistemic terms (e.g., evidence that’s misleading), we should see it as fundamentally pragmatic. In the special case where a disinformant wants to influence another agent's inquiries, disinformation is content that’s used to minimise the probability that the inquisitive agent successfully draws conclusions that threaten the disinformant's interests. This includes misleading first-order evidence that leads agents away from the truth (per Simion's account), but it also includes non-misleading higher-order evidence that accurately shows evidential resources are corrupted, functioning as a deterrent that rationalises withdrawing from dangerous lines of inquiry. (shrink)

A theory of lies should be able to explain, first, what makes lies special among the variety of deceptive uses of language and, second, why we don't consider figurative speakers to be lying even if they seem to be putting forward something they believe to be false. Here I contend that a commitment-based account can meet both aims without compromising with a too narrow conception of lies. The result is an approach that can accommodate the irregular results of experimental analysis, (...) and that answers to the challenge posed by insincere implicatures in general, and lies via figurative language in particular. (shrink)

The inferentialism due to Robert Brandom presents a compelling normative-deontic picture of language and discursive practices, and as such it is well positioned to address phenomena like lying. This short work outlines a simple account of how lying can be conceptualized within that framework. To that end, the basic Brandomian position is extended to include a novel type of status – namely, pseudo-commitments, which are unique in their being non-binding. The traditional definition of lying is then given a status-oriented form, (...) such that lying consists in presenting a pseudo-commitment with the intention for it to be attributed as a standard commitment. Importantly, this can be articulated in our scorekeeping practices, which is demonstrated with three example exchanges – one featuring an undetected lie, one featuring a detected lie, and one featuring a bald-faced lie. Though simple, the account works and invites extension to other similar phenomena, like sarcasm, acting, bullshit, etc. The notion of pseudo-commitment may also find application outside of the inferentialist context. (shrink)