According to an orthodox view, the capacity for conscious experience (sentience) is relevant to the distribution of moral status and value. However, physicalism about consciousness might threaten the normative relevance of sentience. According to the indeterminacy argument, sentience is metaphysically indeterminate while indeterminacy of sentience is incompatible with its normative relevance. According to the introspective argument (by François Kammerer), the unreliability of our conscious introspection undercuts the justification for belief in the normative relevance of consciousness. I defend the normative relevance (...) of sentience against these objections. First, I demonstrate that physicalists only have to concede a limited amount of indeterminacy of sentience. This moderate indeterminacy is in harmony with the role of sentience in determining moral status. Second, I argue that physicalism gives us no reason to expect that introspection is unreliable with respect to the normative relevance of consciousness. (shrink)

Which artificial intelligence (AI) systems are agents? To answer this question, I propose a multidimensional account of agency. According to this account, a system's agency profile is jointly determined by its level of goal-directedness and autonomy as well as is abilities for directly impacting the surrounding world, long-term planning and acting for reasons. Rooted in extant theories of agency, this account enables fine-grained, nuanced comparative characterizations of artificial agency. I show that this account has multiple important virtues and is more (...) informative than alternatives. More speculatively, it may help to illuminate two important emerging questions in AI ethics: 1. Can agency contribute to the moral status of non-human beings, and how? 2. When and why might AI systems exhibit power-seeking behaviour and does this pose an existential risk to humanity? (shrink)

Many researchers and intellectuals warn about extreme risks from artificial intelligence. However, these warnings typically came without systematic arguments in support. This paper provides an argument that AI will lead to the permanent disempowerment of humanity, e.g. human extinction, by 2100. It rests on four substantive premises which it motivates and defends: first, the speed of advances in AI capability, as well as the capability level current systems have already reached, suggest that it is practically possible to build AI systems (...) capable of disempowering humanity by 2100. Second, due to incentives and coordination problems, if it is possible to build such AI, it will be built. Third, since it appears to be a hard technical problem to build AI which is aligned with the goals of its designers, and many actors might build powerful AI, misaligned powerful AI will be built. Fourth, because disempowering humanity is useful for a large range of misaligned goals, such AI will try to disempower humanity. If AI is capable of disempowering humanity and tries to disempower humanity by 2100, then humanity will be disempowered by 2100. This conclusion has immense moral and prudential significance. (shrink)

How can one build AI systems such that they pursue the goals their designers want them to pursue? This is the alignment problem. Numerous authors have raised concerns that, as research advances and systems become more powerful over time, misalignment might lead to catastrophic outcomes, perhaps even to the extinction or permanent disempowerment of humanity. In this paper, I analyze the severity of this risk based on current instances of misalignment. More specifically, I argue that contemporary large language models and (...) game-playing agents are sometimes misaligned. These cases suggest that misalignment tends to have a variety of features: misalignment can be hard to detect, predict and remedy, it does not depend on a specific architecture or training paradigm, it tends to diminish a system’s usefulness and it is the default outcome of creating AI via machine learning. Subsequently, based on these features, I show that the risk of AI alignment magnifies with respect to more capable systems. Not only might more capable systems cause more harm when misaligned, aligning them should be expected to be more difficult than aligning current AI. (shrink)

1. 1.1. Suffering is bad. This is why, ceteris paribus, there are strong moral reasons to prevent suffering. Moreover, typically, those moral reasons are stronger when the amount of suffering at st...

Numerous authors have expressed concern that advanced artificial intelligence (AI) poses an existential risk to humanity. These authors argue that we might build AI which is vastly intellectually superior to humans (a ‘superintelligence’), and which optimizes for goals that strike us as morally bad, or even irrational. Thus, this argument assumes that a superintelligence might have morally bad goals. However, according to some views, a superintelligence necessarily has morally adequate goals. This might be the case either because abilities for moral (...) reasoning and intelligence mutually depend on each other, or because moral realism and moral internalism are true. I argue that the former argument misconstrues the view that intelligence and goals are independent, and that the latter argument misunderstands the implications of moral internalism. Moreover, the current state of AI research provides additional reasons to think that a superintelligence could have bad goals. (shrink)

If a machine attains consciousness, how could we find out? In this paper, I make three related claims regarding positive tests of machine consciousness. All three claims center on the idea that an AI can be constructed “ad hoc”, that is, with the purpose of satisfying a particular test of consciousness while clearly not being conscious. First, a proposed test of machine consciousness can be legitimate, even if AI can be constructed ad hoc specifically to pass this test. This is (...) underscored by the observation that many, if not all, putative tests of machine consciousness can be passed by non-conscious machines via ad hoc means. Second, we can identify ad hoc AI by taking inspiration from the notion of an ad hoc hypothesis in philosophy of science. Third, given the first and the second claim, the most reliable tests of animal consciousness turn out to be valid and useful positive tests of machine consciousness as well. If a non-ad hoc AI exhibits clusters of cognitive capacities facilitated by consciousness in humans which can be selectively switched off by masking and if it reproduces human behavior in suitably designed double dissociation tasks, we should treat the AI as conscious. (shrink)

According to a growing number of researchers, AI may pose catastrophic – or even existential – risks to humanity. Catastrophic risks may be taken to be risks of 100 million human deaths, or a similarly bad outcome. I argue that such risks – while contested – are sufficiently likely to demand rigorous discussion of potential societal responses. Subsequently, I propose four desiderata for approaches to the reduction of catastrophic risks from AI. The quality of such approaches can be assessed by (...) their chance of success, degree of beneficence, degree of non-maleficence, and beneficent side effects. Then, I employ these desiderata to evaluate the promises, limitations and risks of alignment research, timelines research, policy research, halting or slowing down AI research, and compute governance for tackling catastrophic AI risks. While more research is needed, this investigation shows that several approaches for dealing with catastrophic AI risks are available, and where their respective strengths and weaknesses lie. It turns out that many approaches are complementary and that the approaches have a nuanced relationship to approaches to present AI harms. While some approaches are similarly useful for addressing catastrophic risks and present harms, this is not always the case. (shrink)

Clues in the Gospels, evidence from Jewish historian Josephus, belief in the transmigration of souls, and well-documented examples of erroneous declarations of death, combine to support a natural explanation for the Easter story: Jesus survives his short stay on the cross, and is discovered to be barely alive by the few followers who retrieve him. Fearful because they have illegally retrieved a condemned man, they carry out a decoy burial in a tomb. Jesus expires soon after, and is buried quietly (...) in an anonymous grave, but rumor of his survival reaches his followers, as well as the Romans, who open the tomb and find it empty, except for burial linen used in the decoy. To sooth their grief the disciples seize on the rumor of Jesus’ survival and encourage each other to hear the voice and see the image of their master in others. Although the probability of the scenario proposed herein is low, it is surely far larger than the supernatural one. It also is arguably more likely than other non-supernatural hypotheses. It does not require Jesus to survive his crucifixion and meet his followers, or, if Jesus died, to have his body stolen, or be removed from the tomb after the Sabbath and reburied in a lower class graveyard. One only need propose that Jesus was still alive when taken down from the cross and envision how events most likely would unfold from there. (shrink)

Les parlers poitevins nord-occidentaux (Noirmoutier, Marais nord vendéen) présentent une variation allophonique complexe du vocalisme. On peut distinguer plusieurs niveaux de diphtongaison qui rendent ces variétés particulièrement intéressantes pour l’analyse phonologique. L’étonnante diversité des formes phonétiques en surface peut cependant se réduire à deux grandes catégories de noyaux vocaliques, simples (monophtongues) et complexes (monophtongues longues et diphtongues sous-jacentes). Les premières sont sujettes à des contraintes d’expression liées à l’atérité, ou laxité, tandis que les deuxièmes alternent des voyelles tendues avec des (...) diphtongues décroissantes ou fermantes dans l’oral spontané. Cette contribution traite principalement de comment une enquête sur la structure phonologique d’une variété dialectale peut durer vingt ans, en faisant alterner diverses approches et divers angles de vue sur les données : transcription phonétique, spectrographie acoustique, analyse phonologique tendant à devenir une enquête phonologique amenée à remettre en cause ses prémisses et ses instruments d’observation des faits de parole pour aboutir à une description de la langue. (shrink)

L'esprit des Lumières du xvii-xviiie siècle, celui de la raison, de la science, de l'humanisme et du progrès - par opposition à l'obscurantisme, a conduit à l'émergence de connaissances scientifiques disruptives - même en ce qui concerne les fondements de la Société - et indiscutables pour quiconque fait appel à sa raison. Il a ainsi induit de nombreuses transformations dans tous les secteurs de la Société.Aujourd'hui, les technologies numériques, par l'observation de phénomènes invisibles - qui seraient impossibles sans elles - (...) ont déjà induit une multitude de pratiques perturbatrices. Comme cet effet et ses inévitables implications continuent de s'accélérer dans le futur, leur intégration dans la progression de la Société et de l'entreprise est maintenant absolument impérative.Ce livre a pour objet de faire ressortir l'intelligence de la progression de la Société ou de l'entreprise en permettant à l'intelligence du vivre ensemble (Sciences humaines), à l'intelligence des solutions (Sciences exactes) et à l'intelligence de l'artificiel (Sciences de l'ingénieur, y compris le numérique), de se relier entre elles par l'intelligence informationnelle des services. Une telle connexion est établie grâce à la science des services. Elle constitue la base des Lumières informationnelles. (shrink)

L’atlas linguistique finnois de Lauri Kettunen (1940), accessible en ligne, a été initialement conçu par son auteur en fonction de variables de phonologie diachronique. Cependant, en raison de l’intrication de la phonologie dans la morphologie flexionnelle nominale et verbale du finnois, ces données se prêtent aisément à une lecture en termes de taxinomie morphologique. Le finnois apparaît alors comme bien moins « agglutinant » sur le plan typologique, et bien plus inférentiel, ou de type fusionnel. Nous appliquons le modèle PFM (...) (Paradigm Function Morphology) dans cette perspective, en proposant une modélisation diasystémique des classes flexionnelles du finnois. L’approche diachronique de Kettunen, fondée sur des étymons et des schèmes concaténatifs « agglutinants » invite par ailleurs à remonter plus haut dans l’histoire de la langue. On voit alors, comme d’un système somme toute simple de bipartition entre radicaux thématiques et athématiques, de disyllabes et d’imparisyllabiques, le diasystème finnois a évolué vers une complexification croissante, tout en préservant les structures fondamentales héritées. (shrink)