Most epistemologists hold that knowledge entails belief. However, proponents of this claim rarely offer a positive argument in support of it. Rather, they tend to treat the view as obvious and assert that there are no convincing counterexamples. We find this strategy to be problematic. We do not find the standard view obvious, and moreover, we think there are cases in which it is intuitively plausible that a subject knows some proposition P without—or at least without determinately—believing that P. Accordingly, (...) we present five plausible examples of knowledge without (determinate) belief, and we present empirical evidence suggesting that our intuitions about these scenarios are not atypical. (shrink)

We propose four policies of ethical design of human-grade Artificial Intelligence. Two of our policies are precautionary. Given substantial uncertainty both about ethical theory and about the conditions under which AI would have conscious experiences, we should be cautious in our handling of cases where different moral theories or different theories of consciousness would produce very different ethical recommendations. Two of our policies concern respect and freedom. If we design AI that deserves moral consideration equivalent to that of human beings, (...) that AI should be designed with self-respect and with the freedom to explore values other than those we might impose. We are especially concerned about the temptation to create human-grade AI pre-installed with the desire to cheerfully sacrifice itself for its creators’ benefit. (shrink)

Do university ethics classes influence students’ real-world moral choices? We aimed to conduct the first controlled study of the effects of ordinary philosophical ethics classes on real-world moral choices, using non-self-report, non-laboratory behavior as the dependent measure. We assigned 1332 students in four large philosophy classes to either an experimental group on the ethics of eating meat or a control group on the ethics of charitable giving. Students in each group read a philosophy article on their assigned topic and optionally (...) viewed a related video, then met with teaching assistants for 50-minute group discussion sections. They expressed their opinions about meat ethics and charitable giving in a follow-up questionnaire (1032 respondents after exclusions). We obtained 13,642 food purchase receipts from campus restaurants for 495 of the students, before and after the intervention. Purchase of meat products declined in the experimental group (52% of purchases of at least $4.99 contained meat before the intervention, compared to 45% after) but remained the same in the control group (52% both before and after). Ethical opinion also differed, with 43% of students in the experimental group agreeing that eating the meat of factory farmed animals is unethical compared to 29% in the control group. We also attempted to measure food choice using vouchers, but voucher redemption rates were low and no effect was statistically detectable. It remains unclear what aspect of instruction influenced behavior. (shrink)

On an intellectualist approach to belief, the intellectual endorsement of a proposition (such as “The working poor deserve as much respect as the handsomely paid”) is sufficient or nearly sufficient for believing it. On a pragmatic approach to belief, intellectual endorsement is not enough. Belief is behaviorally demanding. To really, fully believe, you must also “walk the walk.” This chapter argues that the pragmatic approach is preferable on pragmatic grounds: It rightly directs our attention to what matters most in thinking (...) about belief. (shrink)

Susan Schneider (2019) has proposed two new tests for consciousness in AI (artificial intelligence) systems, the AI Consciousness Test and the Chip Test. On their face, the two tests seem to have the virtue of proving satisfactory to a wide range of consciousness theorists holding divergent theoretical positions, rather than narrowly relying on the truth of any particular theory of consciousness. Unfortunately, both tests are undermined in having an ‘audience problem’: Those theorists with the kind of architectural worries that motivate (...) the need for such tests should, on similar grounds, doubt that the tests establish the existence of genuine consciousness in the AI in question. Nonetheless, the proposed tests constitute progress, as they could find use by some theorists holding fitting views about consciousness and perhaps in conjunction with other tests for AI consciousness. (shrink)

The mainstream tradition in recent Anglophone Zhuangzi interpretation treats spontaneous skillful responsiveness – similar to the spontaneous responsiveness of a skilled artisan, athlete, or musician – as a, or the, Zhuangzian ideal. However, this interpretation is poorly grounded in the Inner Chapters. On the contrary, in the Inner Chapters, this sort of skillfulness is at least as commonly criticized as celebrated. Even the famous passage about the ox-carving cook might be interpreted more as a celebration of the knife’s passivity than (...) as a celebration of the cook’s skillfulness. (shrink)

Rationalization, in our intended sense of the term, occurs when a person favors a particular conclusion as a result of some factor (such as self-interest) that is of little justificatory epistemic relevance, if that factor then biases the person’s subsequent search for, and assessment of, potential justifications for the conclusion. Empirical evidence suggests that rationalization is common in people’s moral and philosophical thought. We argue that it is likely that the moral and philosophical thought of philosophers and moral psychologists is (...) also pervaded by rationalization. Moreover, although rationalization has some benefits, overall it would be epistemically better if the moral and philosophical reasoning of people, including professional academics, were not as heavily influenced by rationalization as it likely is. We discuss the significance of our arguments for cognitive management and epistemic responsibility. (shrink)

Can we build machines with which we can have interesting conversations? Observing the new optimism of AI regarding deep learning and new language models, we set ourselves an ambitious goal: We want to find out how far we can get in creating a digital replica of a philosopher. This project has two aims; one more technical, investigating of how the best model can be built. The other one, more philosophical, explores the limits and risks which are accompanied by the creation (...) of digital replicas. In cooperation with Daniel Dennett, we took his complete works to fine-tune a GPT-3. In this paper, we share our first results with some piloting models and discuss legal and ethical questions about creating and interacting with digital replicas. (shrink)

An Artificially Intelligent system (an AI) has debatable moral personhood if it is epistemically possible either that the AI is a moral person or that it falls far short of personhood. Debatable moral personhood is a likely outcome of AI development and might arise soon. Debatable AI personhood throws us into a catastrophic moral dilemma: Either treat the systems as moral persons and risk sacrificing real human interests for the sake of entities without interests worth the sacrifice, or do not (...) treat the systems as moral persons and risk perpetrating grievous moral wrongs against them. The moral issues become even more perplexing if we consider cases of possibly conscious AI that are subhuman, superhuman, or highly divergent from us in their morally relevant properties. (shrink)

To study conscious experience we must, to some extent, trust introspective reports; yet introspective reports often do not merit our trust. A century ago, E.B. Titchener advocated extensive introspective training as a means of resolving this difficulty. He describes many of his training techniques in his four-volume laboratory manual of 1901- 1905. This paper explores Titchener's laboratory manual with an eye to general questions about the prospects of introspective training for contemporary consciousness studies, with a focus on the following examples: (...) introspective knowledge of the combination tones that arise when a musical interval is played; the 'flight of colours' in the afterimage of a field of bright, broad- spectrum light; and the possibility of non-obvious visual illusions. Introspective training appears to have some merit, but also to involve significant hazards. (shrink)

One relatively neglected challenge in ethical artificial intelligence (AI) design is ensuring that AI systems invite a degree of emotional and moral concern appropriate to their moral standing. Although experts generally agree that current AI chatbots are not sentient to any meaningful degree, these systems can already provoke substantial attachment and sometimes intense emotional responses in users. Furthermore, rapid advances in AI technology could soon create AIs of plausibly debatable sentience and moral standing, at least by some relevant definitions. Morally (...) confusing AI systems create unfortunate ethical dilemmas for the owners and users of those systems, since it is unclear how those systems ethically should be treated. I argue here that, to the extent possible, we should avoid creating AI systems whose sentience or moral standing is unclear and that AI systems should be designed so as to invite appropriate emotional responses in ordinary users. (shrink)

Philosophers, psychologists, economists and other social scientists continue to debate the nature of human well-being. We argue that this debate centers around five main conceptualizations of well-being: hedonic well-being, life satisfaction, desire fulfillment, eudaimonia, and non-eudaimonic objective-list well-being. Each type of well-being is conceptually different, but are they empirically distinguishable? To address this question, we first developed and validated a measure of desire fulfillment, as no measure existed, and then examined associations between this new measure and several other well-being measures. (...) In addition, we explored associations among all five types of well-being. We found high correlations among all measures of well-being, but generally correlations did not approach unity, even when correcting for unreliability. Furthermore, correlations between well-being and related constructs (e.g., demographics, personality) depended on the type of well-being measured. We conclude that empirical findings based on one type of well-being measure may not generalize to all types of well-being. (shrink)