Can we make machines that think and act like humans or other natural intelligent agents? The answer to this question depends on how we see ourselves and how we see the machines in question. Classical AI and cognitive science had claimed that cognition is computation, and can thus be reproduced on other computing machines, possibly surpassing the abilities of human intelligence. This consensus has now come under threat and the agenda for the philosophy and theory of AI must be set anew, re-defining the relation between AI and Cognitive Science. We can re-claim the original vision of general AI from the technical AI disciplines; we can reject classical cognitive science and replace it with a new theory (e.g. embodied); or we can try to find new ways to approach AI, for example from neuroscience or from systems theory. To do this, we must go back to the basic questions on computing, cognition and ethics for AI. The 30 papers in this volume provide cutting-edge work from leading researchers that define where we stand and where we should go from here.
Ever since the early days of Artificial Intelligence (AI), the complexity of its relationship with philosophy has been under observation. Some devoted their efforts to a systematic foundation of philosophy of AI, taking for granted its placement within philosophy of science. Such endeavors were based on the view of AI as a scientific discipline, primarily aimed at answering questions about the nature of intelligence. Thus, it was natural to consider philosophy of AI, like philosophy of physics and of biology, as part of philosophy of science. We believe, however, that this position must be reconsidered today in the light of the issues recently tackled by AI and of the emergence of new fields of analysis: philosophy of technology and philosophy and engineering. In this paper we analyze how the view of AI as engineering influences philosophy of AI. Moreover, we argue that philosophy of AI, under this influence, can contribute to the foundation of the emerging philosophy of engineering.
For those who find Dreyfus’s critique of AI compelling, the prospects for producing true artificial human intelligence are bleak. An important question thus becomes, what are the prospects for producing artificial non-human intelligence? Applying Dreyfus’s work to this question is difficult, however, because his work is so thoroughly human-centered. Granting Dreyfus that the body is fundamental to intelligence, how are we to conceive of non-human bodies? In this paper, I argue that bringing Dreyfus’s work into conversation with the work of Mark Bickhard offers a way of answering this question, and I try to suggest what doing so means for AI research.
The paper seeks to explore the theoretical foundations as well as lived experience of the users in ambient intelligence. The paper traces the journey of AmI from a ready-to-hand technology to background condition and makes observations regarding its key features viz. physical disappearance and anticipatory responding on the way and shows how this has implications for user as well as environment in the end. AmI is aimed to achieve its transparency by physically disappearing into the environment. In this context, it is argued that it is rather its ability to pervade into those forms of behavior whereby the user accesses her world, i.e. through body and presence, rather than its infrastructural invisibility. The former rather blocks user’s hermeneutic access to it and thereby pushes her to the periphery of her techno-environment. The proper way should be thus allowing AmI to gradually seep into concernful activities of user via learning its present-at-hand features, concealed effectively at present.
The aim of this manuscript is to introduce the notion of experion. This notion is proposed as the primal cognitive unit of neural processing. The proposal focuses on the fact that neural systems have evolved to characterize and act in the situation in which they are involved according to the needs and state of the system, primed by past experience and biased by neurobiological predispositions. The proposal goes on to acknowledge a cluster of principles that characterize neural functioning by its cognitive openness, contingent specialization and selection, as well as cross-modality and heterarchical processing. The proposed framework assumes these facts and hypothesizes that the basic unit is a neural event that holistically integrates all neural processes that take part in addressing the adaptive topic at issue. In particular, I have defined an experion as a neural controlled event within which a particular neuroenvironmental configuration of contents are created to deal with the individual’s adaptive topic at issue. The specific nature of such contents and its ability to address the topic at issue are a product of the deployment of the relevant associations with previous registers of such couplings channeled through the basic operations of the neural architecture. The evolutionary bottom line is that the neural system should not be seen as a system that represents reality, but a system that adapts to it, adjusting the agent to the environment in the best way to obtain its objectives: experiencing, and learning from it.
AI has progressed less than other fields of information technology due to a conceptual impasse. Though much effort has been employed to overcome this situation, often it has been from a restricted point-of-view e.g. philosophy alone or algorithms alone. This paper argues for (and exemplifies) an inter-disciplinary tactic for advancing the field of AI that integrates introspection with programming. The paper has two parts: The first outlines an introspective approach that has been largely overlooked and answers some of the (rather heated) arguments that have caused introspection to be sidelined. The second part offers a practical application of this approach - presented as an algorithm.
There is no doubt that AI research has made significant progress, both in helping us understand how the human mind works and in constructing ever more sophisticated machines. But, for all this, its conceptual foundations remain remarkably unclear and even unsound. In this paper, I take a fresh look, first at the context in which agents must function and so how they must act, and second, at how it is possible for agents to communicate, store and recognise (sensory) messages. This analysis allows a principled distinction to be drawn between the symbolic and connectionist paradigms, showing them to be genuine design alternatives. Further consideration of the connectionist approach seems to offer a number of interesting clues as to how the human brain—apparently of the connectionist ilk—might actually work its incredible magic.
In this chapter, I argue that some aspects of cognitive phenomena cannot be explained computationally. In the first part, I sketch a mechanistic account of computational explanation that spans multiple levels of organization of cognitive systems. In the second part, I turn my attention to what cannot be explained about cognitive systems in this way. I argue that information-processing mechanisms are indispensable in explanations of cognitive phenomena, and this vindicates the computational explanation of cognition. At the same time, it has to be supplemented with other explanations to make the mechanistic explanation complete, and that naturally leads to explanatory pluralism in cognitive science. The price to pay for pluralism, however, is the abandonment of the traditional autonomy thesis asserting that cognition is independent of implementation details.
The Chinese Room Argument purports to show that ‘syntax is not sufficient for semantics’; an argument which led John Searle to conclude that ‘programs are not minds’ and hence that no computational device can ever exhibit true understanding. Yet, although this controversial argument has received a series of criticisms, it has withstood all attempts at decisive rebuttal so far. One of the classical responses to CRA has been based on equipping a purely computational device with a physical robot body. This response, although partially addressed in one of Searle’s original contra arguments - the ‘robot reply’ - more recently gained friction with the development of embodiment and enactivism, two novel approaches to cognitive science that have been exciting roboticists and philosophers alike. Furthermore, recent technological advances - blending biological beings with computational systems - have started to be developed which superficially suggest that mind may be instantiated in computing devices after all. This paper will argue that (a) embodiment alone does not provide any leverage for cognitive robotics wrt the CRA, when based on a weak form of embodiment and that (b) unless they take the body into account seriously, hybrid bio-computer devices will also share the fate of their disembodied or robotic predecessors in failing to escape from Searle’s Chinese room.
Machine ethics and robot rights are quickly becoming hot topics in artificial intelligence/robotics communities. We will argue that the attempts to allow machines to make ethical decisions or to have rights are misguided. Instead we propose a new science of safety engineering for intelligent artificial agents. In particular we issue a challenge to the scientific community to develop intelligent systems capable of proving that they are in fact safe even under recursive self-improvement.
Viewed in the light of the remarkable performance of ‘Watson’ - IBMs proprietary artificial intelligence computer system capable of answering questions posed in natural language - on the US general knowledge quiz show ‘Jeopardy’, we review two experiments on formal systems - one in the domain of quantum physics, the other involving a pictographic languaging game - whereby behaviour seemingly characteristic of domain understanding is generated by the mere mechanical application of simple rules. By re-examining both experiments in the context of Searle’s Chinese Room Argument, we suggest their results merely endorse Searle’s core intuition: that ‘syntactical manipulation of symbols is not sufficient for semantics’. Although, pace Watson, some artificial intelligence practitioners have suggested that more complex, higher-level operations on formal symbols are required to instantiate understanding in computational systems, we show that even high-level calls to Google translate would not enable a computer qua ‘formal symbol processor’ to understand the language it processes. We thus conclude that even the most recent developments in ‘quantum linguistics’ will not enable computational systems to genuinely understand natural language.
Whole brain emulation (WBE) is the possible future one-to-one modeling of the function of the entire (human) brain. The basic idea is to take a particular brain, scan its structure in detail, and construct a software model of it that is so faithful to the original that, when run on appropriate hardware, it will behave in essentially the same way as the original brain. This would achieve software-based intelligence by copying biological intelligence (without necessarily understanding it).
I’ll discuss an interesting argument from the recent book of John Searle Making the Social World (Oxford 2010) that tries to consider the construction of a society as an “engineering” problem and concludes that deontology works against the “computational” or “algorithmic” view of consciousness. I’ll introduce the notion of “consciousness” and the sense in which Searle uses the term (1); I’ll sketch Searle’s argument against the computational model (2) and I’ll criticize Searle’s reasons to warrant his criticism and I try to introduce a “compatibilist” view of human and artificial minds (3).
According to the most popular theories of intentionality, a family of theories we will refer to as “functional intentionality,” a machine can have genuine intentional states so long as it has functionally characterizable mental states that are causally hooked up to the world in the right way. This paper considers a detailed description of a robot that seems to meet the conditions of functional intentionality, but which falls victim to what I call “the composition problem.” One obvious way to escape the problem (arguably, the only way) is if the robot can be shown to be a moral patient – to deserve a particular moral status. If so, it isn’t clear how functional intentionality could remain plausible (something like “phenomenal intentionality” would be required). Finally, while it would have seemed that a reasonable strategy for establishing the moral status of intelligent machines would be to demonstrate that the machine possessed genuine intentionality, the composition argument suggests that the order of precedence is reversed: The machine must first be shown to possess a particular moral status before it is a candidate for having genuine intentionality.
This paper presents some points of proximity between Peirce's insights on the technical/artificial nature of cognition, and contemporary theories of extended cognition. By doing so, it sheds some new light on the possible relevance of Peirce's philosophical approach for artificial intelligence, notably regarding the differences between the reasoning abilities of machines and those of humans.
There is no strong reason to believe human level intelligence represents an upper limit of the capacity of artificial intelligence, should it be realized. This poses serious safety issues, since a superintelligent system would have great power to direct the future according to its possibly flawed goals or motivation systems. Oracle AIs (OAI), confined AIs that can only answer questions, are one particular approach to this problem. However even Oracles are not particularly safe: humans are still vulnerable to traps, social engineering, or simply becoming dependent on the OAI. But OAIs are still strictly safer than general AIs, and there are many extra layers of precautions we can add on top of these. This paper looks at some of them and analyses their strengths and weaknesses.
After a short assessment of the idea behind the Turing Test, its actual status and the overall role it played within AI, I propose a computational cognitive modeling-inspired decomposition of the Turing test as classical “strong AI benchmark” into at least four intermediary testing scenarios: a test for natural language understanding, an evaluation of the performance in emulating human-style rationality, an assessment of creativity-related capacities, and a measure of performance on natural language production of an AI system. I also shortly reflect on advantages and disadvantages of the approach, and conclude with some hints and proposals for further work on the topic.
For at least half a century, it has been popular to compare brains and minds to computers and programs. Despite the continuing appeal of the computational model of the mind, however, it can be difficult to articulate precisely what the view commits one to. Indeed, critics such as John Searle and Hilary Putnam have argued that anything, even a rock, can be viewed as instantiating any computation we please, and this means that the claim that the mind is a computer is not merely false, but it is also deeply confused.
Sensorimotor theories of perception are highly appealing to A.I. due to their apparent simplicity and power; however, they are not problem free either. This paper will presents a frank appraisal of sensorimotor perception discussing and highlighting the good, the bad, and the ugly with respect to a potential sensorimotor A.I.
