Recent research has identified a tension between the Safety principle that knowledge is belief without risk of error, and the Closure principle that knowledge is preserved by competent deduction. Timothy Williamson reconciles Safety and Closure by proposing that when an agent deduces a conclusion from some premises, the agent’s method for believing the conclusion includes their method for believing each premise. We argue that this theory is untenable because it implies problematically easy epistemic access to one’s methods. Several possible solutions (...) are explored and rejected. (shrink)

Under what conditions would an artificially intelligent system have wellbeing? Despite its clear bearing on the ethics of human interactions with artificial systems, this question has received little direct attention. Because all major theories of wellbeing hold that an individual’s welfare level is partially determined by their mental life, we begin by considering whether artificial systems have mental states. We show that a wide range of theories of mental states, when combined with leading theories of wellbeing, predict that certain existing (...) artificial systems have wellbeing. Along the way, we argue that there are good reasons to believe that artificial systems can have wellbeing even if they are not phenomenally conscious. While we do not claim to demonstrate conclusively that AI systems have wellbeing, we argue that there is a significant probability that some AI systems have or will soon have wellbeing, and that this should lead us to reassess our relationship with the intelligent systems we create. (shrink)

We propose developing AIs whose only final goal is being shut down. We argue that this approach to AI safety has three benefits: (i) it could potentially be implemented in reinforcement learning, (ii) it avoids some dangerous instrumental convergence dynamics, and (iii) it creates trip wires for monitoring dangerous capabilities. We also argue that the proposal can overcome a key challenge raised by Soares et al. (2015), that shutdown-seeking AIs will manipulate humans into shutting them down. We conclude by comparing (...) our approach with Soares et al.'s corrigibility framework. (shrink)

I was commissioned by Barry Smith, Editor of The Monist , to act as Advisory Editor for issue 88.1, January 2005 on the topic Humor, and we drafted the appended description. The deadline for submissions is January 31, 2004, and you are welcome to submit an article to me for consideration (word limit 7,500 words, including footnotes). What the Editor and I are, hoping for, is some serious and seriously good philosophical writing on this topic.

Recent advances in natural language processing have given rise to a new kind of AI architecture: the language agent. By repeatedly calling an LLM to perform a variety of cognitive tasks, language agents are able to function autonomously to pursue goals specified in natural language and stored in a human-readable format. Because of their architecture, language agents exhibit behavior that is predictable according to the laws of folk psychology: they function as though they have desires and beliefs, and then make (...) and update plans to pursue their desires given their beliefs. We argue that the rise of language agents significantly reduces the probability of an existential catastrophe due to loss of control over an AGI. This is because the probability of such an existential catastrophe is proportional to the difficulty of aligning AGI systems, and language agents significantly reduce that difficulty. In particular, language agents help to resolve three important issues related to aligning AIs: reward misspecification, goal misgeneralization, and uninterpretability. (shrink)

This paper examines the question of whether Large Language Models (LLMs) like ChatGPT possess minds, focusing specifically on whether they have a genuine folk psychology encompassing beliefs, desires, and intentions. We approach this question by investigating two key aspects: internal representations and dispositions to act. First, we survey various philosophical theories of representation, including informational, causal, structural, and teleosemantic accounts, arguing that LLMs satisfy key conditions proposed by each. We draw on recent interpretability research in machine learning to support these (...) claims. Second, we explore whether LLMs exhibit robust dispositions to perform actions, a necessary component of folk psychology. We consider two prominent philosophical traditions, interpretationism and representationalism, to assess LLM action dispositions. While we find evidence suggesting LLMs may satisfy some criteria for having a mind, particularly in game-theoretic environments, we conclude that the data remains inconclusive. Additionally, we reply to several skeptical challenges to LLM folk psychology, including issues of sensory grounding, the "stochastic parrots" argument, and concerns about memorization. Our paper has three main upshots. First, LLMs do have robust internal representations. Second, there is an open question to answer about whether LLMs have robust action dispositions. Third, existing skeptical challenges to LLM representation do not survive philosophical scrutiny. (shrink)

I will contrast the two main approaches to the foundations of statistical mechanics: the individualist approach and the ensemblist approach. I will indicate the virtues of each, and argue that the conflict between them is perhaps not as great as often imagined.

It is generally assumed that existing artificial systems are not phenomenally conscious, and that the construction of phenomenally conscious artificial systems would require significant technological progress if it is possible at all. We challenge this assumption by arguing that if Global Workspace Theory (GWT) — a leading scientific theory of phenomenal consciousness — is correct, then instances of one widely implemented AI architecture, the artificial language agent, might easily be made phenomenally conscious if they are not already. Along the way, (...) we articulate an explicit methodology for thinking about how to apply scientific theories of consciousness to artificial systems and employ this methodology to arrive at a set of necessary and sufficient conditions for phenomenal consciousness according to GWT. (shrink)

A syntactically correct number-specification may fail to specify any number due to underspecification. For similar reasons, although each sentence in the Yablo sequence is syntactically perfect, none yields a statement with any truth-value. As is true of all members of the Liar family, the sentences in the Yablo sequence are so constructed that the specification of their truth-conditions is vacuous; the Yablo sentences fail to yield statements. The ‘revenge’ problem is easily defused. The solution to the semantical paradoxes offered here (...) revives the mediaeval cassatio approach, one that largely disappeared due to its incomprehending rejection by influential contemporary writers such as William Shyreswood and Thomas Bradwardine. The diagnosis readily extends to the set-theoretic paradoxes. (shrink)

Bohmian mechanics is arguably the most naively obvious embedding imaginable of Schr¨ odinger’s equation into a completely coherent physical theory. It describes a world in which particles move in a highly non-Newtonian sort of way, one which may at first appear to have little to do with the spectrum of predictions of quantum mechanics. It turns out, however, that as a consequence of the defining dynamical equations of Bohmian mechanics, when a system has wave function ψ its configuration is typically (...) random, with probability density ρ given by |ψ|2, the quantum equilibrium distribution. It also turns out that the entire quantum formalism, operators as observables and all the rest, naturally emerges in Bohmian mechanics from the analysis of “measurements.” This analysis reveals the status of operators as observables in the description of quantum phenomena, and facilitates a clear view of the range of applicability of the usual quantum mechanical formulas. (shrink)

Triviality results threaten plausible principles governing our credence in epistemic modal claims. This paper develops a new account of modal credence which avoids triviality. On the resulting theory, probabilities are assigned not to sets of worlds, but rather to sets of information state-world pairs. The theory avoids triviality by giving up the principle that rational credence is closed under conditionalization. A rational agent can become irrational by conditionalizing on new evidence. In place of conditionalization, the paper develops a new account (...) of updating: conditionalization with normalization. (shrink)

Laurence Goldstein gives a straightforward and lively account of some of the central themes of Wittgenstein's writings on meaning, mind, and mathematics.

With the rise of the system approach in World War II, US-engineers developed new design techniques to manage complex socio-technical systems, especially military ones. One of these new methods was the use of block diagrams which should guarantee a rational design process without eliminating the traditional visual habitus of the engineering community. This text argues that the block diagram, as a new visual design method, was embedded in the process of establishing a new engineering science, called system engineering. Furthermore, the (...) block diagram symbolized the changed role of technical experts in the Cold War era, who were regarded as competent problem solvers and as a community that could protect the national security by developing complex military technology. Two images, that were parts of advertisments in theScientific American, shall be used to demonstrate the block diagram’s function in the public sphere as a symbol that represents a new technocratic interpretation of modernity. (shrink)

This book is about the mind and its connection to the brain. The first two chapters discuss the basic characteristics of the mind, and places it in historical context by noting trends in popular culture, and various people's ideas about the mind. This discussion ends by concluding that the most fruitful approach to studying the mind is a scientific approach that looks for connections between the mind and the brain. The last four chapters focus on the following specific principles: The (...) operation of the mind is hidden; Prediction is an important function of the mind; and Connectivity and communication between different areas of the mind are crucial for the mind's operation. An important feature of this discussion is that it is supported by empirical research, by briefly describing experiments. This book therefore goes beyond simply presenting conclusions, by elucidating the procedures and reasoning that led to those conclusions. The intent of the book is to give readers a feel for the latest scientific thinking about the mind, by focusing on basic principles that hold across many different functions of the mind. One key to marketing the book is to emphasize the fact that the writing makes a fascinating, but complex, subject accessible to the general reader. (shrink)

In recent years, a number of theorists have claimed that beliefs about probability are transparent. To believe probably p is simply to have a high credence that p. In this paper, I prove a variety of triviality results for theses like the above. I show that such claims are inconsistent with the thesis that probabilistic modal sentences have propositions or sets of worlds as their meaning. Then I consider the extent to which a dynamic semantics for probabilistic modals can capture (...) theses connecting belief, certainty, credence, and probability. I show that although a dynamic semantics for probabilistic modals does allow one to validate such theses, it can only do so at a cost. I prove that such theses can only be valid if probabilistic modals do not satisfy the axioms of the probability calculus. (shrink)

Foreword by Oliver Sacks Kurt Goldstein (1878-1965) was already an established neuropsychologist when he emigrated from Germany to the United States in the 1930s. This book, his magnum opus and widely regarded as a modern classic in psychology and biology, grew out of his dissatisfaction with traditional natural science techniques for analyzing living beings. It offers a broad introduction to the sources and ranges of application of the "holistic" or "organismic" research program that has since become a standard part (...) of biological thought. Goldstein was especially concerned with the breakdown of organization and the failure of central controls that take place in catastrophic responses to situations such as physical or mental illness. But he was equally attuned to the amazing powers of the organism to readjust to such catastrophic losses, if only by withdrawal to a more limited range that it could manage by a redistribution of its reduced energies, thus reclaiming as much wholeness as new circumstances allowed. Goldstein's theses in The Organism have had an important impact on philosophical and psychological thought throughout the twentieth century, as evidenced in the work of Maurice Merleau-Ponty, Georges Canguilhem, Ernst Cassirer, and Ludwig Binswanger. (shrink)

[Correction Notice: An erratum for this article was reported in Vol 109 of Psychological Review. Due to circumstances that were beyond the control of the authors, the studies reported in "Models of Ecological Rationality: The Recognition Heuristic," by Daniel G. Goldstein and Gerd Gigerenzer overlap with studies reported in "The Recognition Heuristic: How Ignorance Makes Us Smart," by the same authors and with studies reported in "Inference From Ignorance: The Recognition Heuristic". In addition, Figure 3 in the Psychological Review (...) article was originally published in the book chapter and should have carried a note saying that it was used by permission of Oxford University Press.] One view of heuristics is that they are imperfect versions of optimal statistical procedures considered too complicated for ordinary minds to carry out. In contrast, the authors consider heuristics to be adaptive strategies that evolved in tandem with fundamental psychological mechanisms. The recognition heuristic, arguably the most frugal of all heuristics, makes inferences from patterns of missing knowledge. This heuristic exploits a fundamental adaptation of many organisms: the vast, sensitive, and reliable capacity for recognition. The authors specify the conditions under which the recognition heuristic is successful and when it leads to the counter-intuitive less-is-more effect in which less knowledge is better than more for making accurate inferences. (shrink)

The most puzzling issue in the foundations of quantum mechanics is perhaps that of the status of the wave function of a system in a quantum universe. Is the wave function objective or subjective? Does it represent the physical state of the system or merely our information about the system? And if the former, does it provide a complete description of the system or only a partial description? We shall address these questions here mainly from a Bohmian perspective, and shall (...) argue that part of the difficulty in ascertaining the status of the wave function in quantum mechanics arises from the fact that there are two different sorts of wave functions involved. The most fundamental wave function is that of the universe. From it, together with the configuration of the universe, one can define the wave function of a subsystem. We argue that the fundamental wave function, the wave function of the universe, has a law-like character. (shrink)

You omega know something when you know it, and know that you know it, and know that you know that you know it, and so on. This paper first argues that omega knowledge matters, in the sense that it is required for rational assertion, action, inquiry, and belief. The paper argues that existing accounts of omega knowledge face major challenges. One account is skeptical, claiming that we have no omega knowledge of any ordinary claims about the world. Another account embraces (...) the KK thesis, and identifies knowledge with omega knowledge. This position faces counterexamples, and struggles to make sense of inexact knowledge. The paper then develops a new account of knowledge, by proposing the principle of Reflective Luminosity: if you know that you know something, then you omega know it. I argue that Reflective Luminosity allows for omega knowledge while avoiding the problems for KK. (shrink)

This paper presents a direction for narrative ethics based on ethical ideas found in the works of Michel Foucault. Narrative ethics is understood here at the meta-level of cultural discourse to see how the moral subject is constituted by the discursive practices that structure the contemporary debate on reproductive technologies. At this level it becomes meta-narrative-ethics. After a theoretical discussion, this paper uses two literary narratives representing the polarized views in the debate to show how the moral subject may be (...) compelled to relate to its self. Ethics is redefined as Foucauldian rapport à soi, and ethical analysis, at this meta-level, shows how the moral self is intimately connected to cultural discourse. (shrink)

Context: Political psychology rests on the assumption of the existence of a world outside and independent of consciousness. This ontological materialism is hardly spoken of within the field, as it ….

In [3] John S. Bell proposed how to associate particle trajectories with a lattice quantum field theory, yielding what can be regarded as a |Ψ|2-distributed Markov process on the appropriate configuration space. A similar process can be defined in the continuum, for more or less any regularized quantum field theory; such processes we call Bell-type quantum field theories. We describe methods for explicitly constructing these processes. These concern, in addition to the definition of the Markov processes, the efficient calculation of (...) jump rates, how to obtain the process from the processes corresponding to the free and interaction Hamiltonian alone, and how to obtain the free process from the free Hamiltonian or, alternatively, from the one-particle process by a construction analogous to “second quantization.” As an example, we consider the process for a second quantized Dirac field in an external electromagnetic field. (shrink)

LLMs have dramatically improved in capabilities in recent years. This raises the question of whether LLMs could become genuine agents with beliefs and desires. This paper demonstrates an in principle limit to LLM agency, based on their architecture. LLMs are next word predictors: given a string of text, they calculate the probability that various words can come next. LLMs produce outputs that reflect these probabilities. I show that next word predictors are exploitable. If LLMs are prompted to make probabilistic predictions (...) about the world, these predictions are guaranteed to be incoherent, and so Dutch bookable. If LLMs are prompted to make choices over actions, their preferences are guaranteed to be intransitive, and so money pumpable. In short, the problem is that selecting an action based on its potential value is structurally different then selecting the description of an action that is most likely given a prompt: probability cannot be forced into the shape of expected value. The in principle exploitability of LLMs raises doubts about how agential they can become. This exploitability also offers an opportunity for humanity to safely control such AI systems. (shrink)

This paper offers the first careful analysis of the possibility that AI and humanity will go to war. The paper focuses on the case of artificial general intelligence, AI with broadly human capabilities. The paper uses a bargaining model of war to apply standard causes of war to the special case of AI/human conflict. The paper argues that information failures and commitment problems are especially likely in AI/human conflict. Information failures would be driven by the difficulty of measuring AI capabilities, (...) by the uninterpretability of AI systems, and by differences in how AIs and humans analyze information. Commitment problems would make it difficult for AIs and humans to strike credible bargains. Commitment problems could arise from power shifts, rapid and discontinuous increases in AI capabilities. Commitment problems could also arise from missing focal points, where AIs and humans fail to effectively coordinate on policies to limit war. In the face of this heightened chance of war, the paper proposes several interventions. War can be made less likely by improving the measurement of AI capabilities, capping improvements in AI capabilities, designing AI systems to be similar to humans, and by allowing AI systems to participate in democratic political institutions. (shrink)

We discuss the content and significance of John von Neumann’s quantum ergodic theorem (QET) of 1929, a strong result arising from the mere mathematical structure of quantum mechanics. The QET is a precise formulation of what we call normal typicality, i.e., the statement that, for typical large systems, every initial wave function ψ0 from an energy shell is “normal”: it evolves in such a way that |ψt ψt| is, for most t, macroscopically equivalent to the micro-canonical density matrix. The QET (...) has been mostly forgotten after it was criticized as a dynamically vacuous statement in several papers in the 1950s. However, we point out that this criticism does not apply to the actual QET, a correct statement of which does not appear in these papers, but to a different (indeed weaker) statement. Furthermore, we formulate a stronger statement of normal typicality, based on the observation that the bound on the deviations from the average specified by von Neumann is unnecessarily coarse and a much tighter (and more relevant) bound actually follows from his proof. (shrink)

Unconscious processes are mental states that occur in the absence of subjective awareness. We offer a focused historical survey of the robust debate about the nature of unconscious mental processing, from ancient and medieval theories that allow for bodily functions without subjective awareness to the 20th century acceptance of autonomous unconscious processing. The background introduction culminates with the rise of cognitive science in the latter half of the 20th century, as dual systems theories claimed that the mind had two forms (...) of processing: an unconscious system that is associative, fast, and automatic and a conscious system that performs sequential, deliberative, and slow mental functions. More recently, the “cognitive unconscious” has been discovered to perform most if not all cognitive functions including higher-level cognitive processing, such as addition and subtraction. Despite a rich history of experimental methods and findings, recent criticisms are forcing scientists to reexamine these advancements. We are optimistic that the field will develop novel techniques that answer these criticisms, including measures for studying subjective reports in the absence of subjective awareness of the content of the unconscious state and the act of reporting. (shrink)

The renewed interest in the foundations of quantum statistical mechanics in recent years has led us to study John von Neumann’s 1929 article on the quantum ergodic theorem. We have found this almost forgotten article, which until now has been available only in German, to be a treasure chest, and to be much misunderstood. In it, von Neumann studied the long-time behavior of macroscopic quantum systems. While one of the two theorems announced in his title, the one he calls the (...) “quantum H-theorem,” is actually a much weaker statement than Boltzmann’s classical H-theorem, the other theorem, which he calls the “quantum ergodic theorem,” is a beautiful and very non-trivial result. It expresses a fact we call “normal typicality” and can be summarized as follows: For a “typical” finite family of commuting macroscopic observables, every initial wave function ψ0 from a micro-canonical energy shell so evolves that for most times t in the long run, the joint probability distribution of these observables obtained from ψ is close to.. (shrink)

Consideration of a paradox originally discovered by John Buridan provides a springboard for a general solution to paradoxes within the Liar family. The solution rests on a philosophical defence of truth-value-gaps and is consistent (non-dialetheist), avoids ‘revenge’ problems, imports no ad hoc assumptions, is not applicable to only a proper subset of the semantic paradoxes and implies no restriction of the expressive capacities of language.

Quality and service improvement (QSI) research employs a broad range of methods to enhance the efficiency of healthcare delivery. QSI research differs from traditional healthcare research and poses unique ethical questions. Since QSI research aims to generate knowledge to enhance quality improvement efforts, should it be considered research for regulatory purposes? Is review by a research ethics committee required? Should healthcare providers be considered research participants? If participation in QSI research entails no more than minimal risk, is consent required? The (...) lack of consensus on answers to these questions highlights the need for ethical guidance. Three distinct approaches to classifying QSI research in accordance with existing ethical principles and regulations can be found in the literature. In the first approach, QSI research is viewed as distinct from other types of healthcare research and does not require regulation. In the second approach, QSI research falls within regulatory guidelines but is exempt from research ethics committee review. In the third approach, QSI research is deemed to be part of the learning healthcare system and, as such, is subject to a different set of ethical principles entirely. In this paper, we critically assess each of these views. While none of these approaches is entirely satisfactory, we argue that use of the ethical principles governing research provides the best means of addressing the numerous questions posed by QSI research. (shrink)

This thesis addresses the tension in pragmatic cluster-randomized trials between their social value and the requirement to respect the autonomy of research participants. Pragmatic trials are designed to evaluate the effectiveness of treatments in real-world settings to inform clinical decision-making and promote cost-efficient care. These trials are often embedded into clinical settings and ideally include all patients who would receive the treatments under investigation as a part of routine care. Trialists increasingly adopt cluster-randomized designs—in which intact groups, such as hospitals (...) or clinics, are allocated randomly to study interventions—to simplify the inclusion of all patients. But including all-comers conflicts with the requirement to obtain written informed consent from research participants. Since informed consent is central to respecting patient autonomy, the question arises: how can the ends of autonomy and pragmatism be served simultaneously? Some philosophers argue that patients have an obligation to participate in clinical research and that this may negate consent requirements. I argue that while there may be grounds for a prima facie obligation for patients to participate in clinical research, no compelling argument has demonstrated that the obligation is enforceable. Others assert that broad application of a waiver of consent will facilitate the conduct of pragmatic cluster-randomized trials. I demonstrate that this proposal sharply conflicts with the historical origins of the waiver. I articulate a novel moral foundation for the use of a waiver of consent and show that when trials evaluate treatments delivered directly to patients, the autonomy interests at stake for participants are too substantial to permit its use. My solution draws a distinction between consent requirements in existing policy and consent as an autonomous authorization. As many pragmatic cluster-randomized trials are conducted in primary care settings with no research staff, I argue that patient autonomy can be promoted and protected using clinical-style consent, in which health providers seek verbal informed consent from patients and document it in the electronic health record. This approach has been associated with high rates of recruitment and, thus, may satisfy both requirements for social value and respect for autonomy. (shrink)

We consider a one-dimensional translation invariant point process of density one with uniformly bounded variance of the number NI of particles in any interval I. Despite this suppression of fluctuations we obtain a large deviation principle with rate function F(ρ) −L−1 log Prob(ρ) for observing a macroscopic density profile ρ(x), x ∈ [0, 1], corresponding to the coarse-grained and rescaled density of the points of the original process in an interval of length L in the limit L → ∞. F(ρ) (...) is not convex and is discontinuous at ρ ≡ 1, the typical profile. (shrink)

Wittgenstein links the strange phenomenon of experiencing meaning to the more familiar phenomenon of seeing-as, or noticing an aspect. His interest in the subject seems to have been sparked by the work of William James, and this chapter examines both what he has to say on the matter (some of which long pre-dates the 'third' Wittgenstein stage) and its relevance to language-learning, prose, poetry and puns.

We consider the possibility that all particles in the world are fundamentally identical, i.e., belong to the same species. Different masses, charges, spins, flavors, or colors then merely correspond to different quantum states of the same particle, just as spin-up and spin-down do. The implications of this viewpoint can be best appreciated within Bohmian mechanics, a precise formulation of quantum mechanics with particle trajectories. The implementation of this viewpoint in such a theory leads to trajectories different from those of the (...) usual formulation, and thus to a version of Bohmian mechanics that is inequivalent to, though arguably empirically indistinguishable from, the usual one. The mathematical core of this viewpoint is however rather independent of the detailed dynamical scheme Bohmian mechanics provides, and it amounts to the assertion that the configuration space for N particles, even N “distinguishable particles,” is the set of all N -point subsets of physical 3-space. (shrink)