The EU Artificial Intelligence Act (AIA) defines four risk categories for AI systems: unacceptable, high, limited, and minimal. However, it lacks a clear methodology for the assessment of these risks in concrete situations. Risks are broadly categorized based on the application areas of AI systems and ambiguous risk factors. This paper suggests a methodology for assessing AI risk magnitudes, focusing on the construction of real-world risk scenarios. To this scope, we propose to integrate the AIA with (...) a framework developed by the Intergovernmental Panel on Climate Change (IPCC) reports and related literature. This approach enables a nuanced analysis of AI risk by exploring the interplay between (a) risk determinants, (b) individual drivers of determinants, and (c) multiple risk types. We further refine the proposed methodology by applying a proportionality test to balance the competing values involved in AI risk assessment. Finally, we present three uses of this approach under the AIA: to implement the Regulation, to assess the significance of risks, and to develop internal risk management systems for AI deployers. (shrink)

The EU proposal for the Artificial Intelligence Act (AIA) defines four risk categories: unacceptable, high, limited, and minimal. However, as these categories statically depend on broad fields of application of AI, the risk magnitude may be wrongly estimated, and the AIA may not be enforced effectively. This problem is particularly challenging when it comes to regulating general-purpose AI (GPAI), which has versatile and often unpredictable applications. Recent amendments to the compromise text, though introducing context-specific assessments, remain insufficient. To (...) address this, we propose applying the risk categories to specific AI scenarios, rather than solely to fields of application, using a risk assessment model that integrates the AIA with the risk approach arising from the Intergovernmental Panel on Climate Change (IPCC) and related literature. This model enables the estimation of the magnitude of AI risk by considering the interaction between (a) risk determinants, (b) individual drivers of determinants, and (c) multiple risk types. We use large language models (LLMs) as an example. (shrink)

In this work, we survey skepticism regarding AI risk and show parallels with other types of scientific skepticism. We start by classifying different types of AI Risk skepticism and analyze their root causes. We conclude by suggesting some intervention approaches, which may be successful in reducing AI risk skepticism, at least amongst artificial intelligence researchers.

This essay argues that a new subfield of AI governance should be explored that examines the policy-making process and its implications for AI governance. A growing number of researchers have begun working on the question of how to mitigate the catastrophic risks of transformative artificial intelligence, including what policies states should adopt. However, this essay identifies a preceding, meta-level problem of how the space of possible policies is affected by the politics and administrative mechanisms of how those policies are created (...) and implemented. This creates a new set of key considerations for the field of AI governance and should influence the action of future policymakers. This essay examines some of the theories of the policymaking process, how they compare to current work in AI governance, and their implications for the field at large and ends by identifying areas of future research. (shrink)

In an effort to inform the discussion surrounding existential risks from AI, we formulate Extinction-level Goodhart’s Law as “Virtually any goal specification, pursued to the extreme, will result in the extinction of humanity”, and we aim to understand which formal models are suitable for investigating this hypothesis. Note that we remain agnostic as to whether Extinction-level Goodhart’s Law holds or not. As our key contribution, we identify a set of conditions that are necessary for a model that aims to be (...) informative for evaluating specific arguments for Extinction-level Goodhart’s Law. Since each of the conditions seems to significantly contribute to the complexity of the resulting model, formally evaluating the hypothesis might be exceedingly difficult. This raises the possibility that whether the risk of extinction from artificial intelligence is real or not, the underlying dynamics might be invisible to current scientific methods. (shrink)

The standard argument to the conclusion that artificial intelligence (AI) constitutes an existential risk for the human species uses two premises: (1) AI may reach superintelligent levels, at which point we humans lose control (the ‘singularity claim’); (2) Any level of intelligence can go along with any goal (the ‘orthogonality thesis’). We find that the singularity claim requires a notion of ‘general intelligence’, while the orthogonality thesis requires a notion of ‘instrumental intelligence’. If this interpretation is correct, they cannot (...) be joined as premises and the argument for the existential risk of AI turns out invalid. If the interpretation is incorrect and both premises use the same notion of intelligence, then at least one of the premises is false and the orthogonality thesis remains itself orthogonal to the argument to existential risk from AI. In either case, the standard argument for existential risk from AI is not sound.—Having said that, there remains a risk of instrumental AI to cause very significant damage if designed or used badly, though this is not due to superintelligence or a singularity. (shrink)

An advanced artificial intelligence could pose a significant existential risk to humanity. Several research institutes have been set-up to address those risks. And there is an increasing number of academic publications analysing and evaluating their seriousness. Nick Bostrom’s superintelligence: paths, dangers, strategies represents the apotheosis of this trend. In this article, I argue that in defending the credibility of AI risk, Bostrom makes an epistemic move that is analogous to one made by so-called sceptical theists in the debate (...) about the existence of God. And while this analogy is interesting in its own right, what is more interesting are its potential implications. It has been repeatedly argued that sceptical theism has devastating effects on our beliefs and practices. Could it be that AI-doomsaying has similar effects? I argue that it could. Specifically, and somewhat paradoxically, I argue that it could amount to either a reductio of the doomsayers position, or an important and additional reason to join their cause. I use this paradox to suggest that the modal standards for argument in the superintelligence debate need to be addressed. (shrink)

An increasing number of financial services (FS) companies are adopting solutions driven by artificial intelligence (AI) to gain operational efficiencies, derive strategic insights, and improve customer engagement. However, the rate of adoption has been low, in part due to the apprehension around its complexity and self-learning capability, which makes auditability a challenge in a highly regulated industry. There is limited literature on how FS companies can implement the governance and controls specific to AI-driven solutions. AI auditing cannot be performed in (...) a vacuum; the risks are not confined to the algorithm itself, but rather permeates the entire organization. Using the risk of unfairness as an example, this paper will introduce the overarching governance strategy and control framework to address the practical challenges in mitigating risks AI introduces. With regulatory implications and industry use cases, this framework will enable leaders to innovate with confidence. (shrink)

This article reports the findings of AI4People, an Atomium—EISMD initiative designed to lay the foundations for a “Good AI Society”. We introduce the core opportunities and risks of AI for society; present a synthesis of five ethical principles that should undergird its development and adoption; and offer 20 concrete recommendations—to assess, to develop, to incentivise, and to support good AI—which in some cases may be undertaken directly by national or supranational policy makers, while in others may be led by other (...) stakeholders. If adopted, these recommendations would serve as a firm foundation for the establishment of a Good AI Society. (shrink)

Abstract: This article examines risks associated with the program of passive search for alien signals (Search for Extraterrestrial Intelligence, or SETI) connected with the possibility of finding of alien transmission which includes description of AI system aimed on self-replication (SETI-attack). A scenario of potential vulnerability is proposed as well as the reasons why the proportion of dangerous to harmless signals may be high. The article identifies necessary conditions for the feasibility and effectiveness of the SETI-attack: ETI existence, possibility of AI, (...) small size of the Seed AI, small speed of physical interstellar travel and large distance between civilizations. Needed additions to the SETI protocol are considered: keep the signal existence, its content and source location in secret, don’t run alien programs, jam dangerous signals, wait until creation of humanity’s own AI. In addition, scenarios in which it may be reasonable to start an alien AI (if it has been downloaded) are explored. (shrink)

This paper presents an overview of the risks and benefits of Social AI, understood as conversational AI systems that cater to human social needs like romance, companionship, or entertainment. Section 1 of the paper provides a brief history of conversational AI systems and introduces conceptual distinctions to help distinguish varieties of Social AI and pathways to their deployment. Section 2 of the paper adds further context via a brief discussion of anthropomorphism and its relevance to assessment of human-chatbot relationships. Section (...) 3 of the paper provides a survey of potential and in some cases demonstrated harms associated with user interactions with Social AI systems. Finally, Section 4 discusses how the benefits and harms of Social AI can best be addressed, with a primary focus on how frameworks from AI ethics can inform their development. (shrink)

Papers from the conference on AI Risk (published in JETAI), supplemented by additional work. --- If the intelligence of artificial systems were to surpass that of humans, humanity would face significant risks. The time has come to consider these issues, and this consideration must include progress in artificial intelligence (AI) as much as insights from AI theory. -- Featuring contributions from leading experts and thinkers in artificial intelligence, Risks of Artificial Intelligence is the first volume of collected chapters dedicated (...) to examining the risks of AI. The book evaluates predictions of the future of AI, proposes ways to ensure that AI systems will be beneficial to humans, and then critically evaluates such proposals. 1 Vincent C. Müller, Editorial: Risks of Artificial Intelligence - 2 Steve Omohundro, Autonomous Technology and the Greater Human Good - 3 Stuart Armstrong, Kaj Sotala and Sean O’Heigeartaigh, The Errors, Insights and Lessons of Famous AI Predictions - and What they Mean for the Future - 4 Ted Goertzel, The Path to More General Artificial Intelligence - 5 Miles Brundage, Limitations and Risks of Machine Ethics - 6 Roman Yampolskiy, Utility Function Security in Artificially Intelligent Agents - 7 Ben Goertzel, GOLEM: Toward an AGI Meta-Architecture Enabling Both Goal Preservation and Radical Self-Improvement - 8 Alexey Potapov and Sergey Rodionov, Universal Empathy and Ethical Bias for Artificial General Intelligence - 9 András Kornai, Bounding the Impact of AGI - 10 Anders Sandberg, Ethics and Impact of Brain Emulations 11 Daniel Dewey, Long-Term Strategies for Ending Existential Risk from Fast Takeoff - 12 Mark Bishop, The Singularity, or How I Learned to Stop Worrying and Love AI -. (shrink)

If the intelligence of artificial systems were to surpass that of humans significantly, this would constitute a significant risk for humanity. Time has come to consider these issues, and this consideration must include progress in AI as much as insights from the theory of AI. The papers in this volume try to make cautious headway in setting the problem, evaluating predictions on the future of AI, proposing ways to ensure that AI systems will be beneficial to humans – and (...) critically evaluating such proposals. (shrink)

In this article we analyse the role that artificial intelligence (AI) could play, and is playing, to combat global climate change. We identify two crucial opportunities that AI offers in this domain: it can help improve and expand current understanding of climate change and it contribute to combating the climate crisis effectively. However, the development of AI also raises two sets of problems when considering climate change: the possible exacerbation of social and ethical challenges already associated with AI, and the (...) contribution to climate change of the greenhouse gases emitted by training data and computation-intensive AI systems. We assess the carbon footprint of AI research, and the factors that influence AI’s greenhouse gas (GHG) emissions in this domain. We find that the carbon footprint of AI research may be significant and highlight the need for more evidence concerning the trade-off between the GHG emissions generated by AI research and the energy and resource efficiency gains that AI can offer. In light of our analysis, we argue that leveraging the opportunities offered by AI for global climate change whilst limiting its risks is a gambit which requires responsive, evidence-based and effective governance to become a winning strategy. We conclude by identifying the European Union as being especially well-placed to play a leading role in this policy response and provide 13 recommendations that are designed to identify and harness the opportunities of AI for combating climate change, while reducing its impact on the environment. (shrink)

A classification of the global catastrophic risks of AI is presented, along with a comprehensive list of previously identified risks. This classification allows the identification of several new risks. We show that at each level of AI’s intelligence power, separate types of possible catastrophes dominate. Our classification demonstrates that the field of AI risks is diverse, and includes many scenarios beyond the commonly discussed cases of a paperclip maximizer or robot-caused unemployment. Global catastrophic failure could happen at various levels of (...) AI development, namely, before it starts self-improvement, during its takeoff, when it uses various instruments to escape its initial confinement, or after it successfully takes over the world and starts to implement its goal system, which could be plainly unaligned, or feature-flawed friendliness. AI could also halt at later stages of its development either due to technical glitches or ontological problems. Overall, we identified around several dozen scenarios of AI-driven global catastrophe. The extent of this list illustrates that there is no one simple solution to the problem of AI safety, and that AI safety theory is complex and must be customized for each AI development level. (shrink)

Recent progress in artificial intelligence (AI) has drawn attention to the technology’s transformative potential, including what some see as its prospects for causing large-scale harm. We review two influential arguments purporting to show how AI could pose catastrophic risks. The first argument — the Problem of Power-Seeking — claims that, under certain assumptions, advanced AI systems are likely to engage in dangerous power-seeking behavior in pursuit of their goals. We review reasons for thinking that AI systems might seek power, that (...) they might obtain it, that this could lead to catastrophe, and that we might build and deploy such systems anyway. The second argument claims that the development of human-level AI will unlock rapid further progress, culminating in AI systems far more capable than any human — this is the Singularity Hypothesis. Power-seeking behavior on the part of such systems might be particularly dangerous. We discuss a variety of objections to both arguments and conclude by assessing the state of the debate. (shrink)

Special Issue “Risks of artificial general intelligence”, Journal of Experimental and Theoretical Artificial Intelligence, 26/3 (2014), ed. Vincent C. Müller. http://www.tandfonline.com/toc/teta20/26/3# - Risks of general artificial intelligence, Vincent C. Müller, pages 297-301 - Autonomous technology and the greater human good - Steve Omohundro - pages 303-315 - - - The errors, insights and lessons of famous AI predictions – and what they mean for the future - Stuart Armstrong, Kaj Sotala & Seán S. Ó hÉigeartaigh - pages 317-342 - - (...) - The path to more general artificial intelligence - Ted Goertzel - pages 343-354 - - - Limitations and risks of machine ethics - Miles Brundage - pages 355-372 - - - Utility function security in artificially intelligent agents - Roman V. Yampolskiy - pages 373-389 - - - GOLEM: towards an AGI meta-architecture enabling both goal preservation and radical self-improvement - Ben Goertzel - pages 391-403 - - - Universal empathy and ethical bias for artificial general intelligence - Alexey Potapov & Sergey Rodionov - pages 405-416 - - - Bounding the impact of AGI - András Kornai - pages 417-438 - - - Ethics of brain emulations - Anders Sandberg - pages 439-457. (shrink)

We are about to see the emergence of religions devoted to the worship of Artificial Intelligence (AI). Such religions pose acute risks, both to their followers and to the public. We should require their creators, and governments, to acknowledge these risks and to manage them as best they can. However, these new religions cannot be stopped altogether, nor should we try to stop them if we could. We must accept that AI worship will become part of our culture, and we (...) must learn to live with it. Indeed, we should do much more than this. We should value it as a new source of meaning in a rapidly-changing world. AI-based religions deserve respect and recognition alongside other, more established ones, and they should be granted the same level of legal protection. (shrink)

One of the primary, if not most critical, difficulties in the design and implementation of autonomous systems is the black-boxed nature of the decision-making structures and logical pathways. How human values are embodied and actualised in situ may ultimately prove to be harmful if not outright recalcitrant. For this reason, the values of stakeholders become of particular significance given the risks posed by opaque structures of intelligent agents (IAs). This paper explores how decision matrix algorithms, via the belief-desire-intention model for (...) autonomous vehicles, can be designed to minimize the risks of opaque architectures. Primarily through an explicit orientation towards designing for the values of explainability and verifiability. In doing so, this research adopts the Value Sensitive Design (VSD) approach as a principled framework for the incorporation of such values within design. VSD is recognized as a potential starting point that offers a systematic way for engineering teams to formally incorporate existing technical solutions within ethical design, while simultaneously remaining pliable to emerging issues and needs. It is concluded that the VSD methodology offers at least a strong enough foundation from which designers can begin to anticipate design needs and formulate salient design flows that can be adapted to the changing ethical landscapes required for utilisation in autonomous vehicles. (shrink)

Deepfakes pose a multi-faceted threat to the acquisition of knowledge. It is widely hoped that technological solutions—in the form of artificially intelligent systems for detecting deepfakes—will help to address this threat. I argue that the prospects for purely technological solutions to the problem of deepfakes are dim. Especially given the evolving nature of the threat, technological solutions cannot be expected to prevent deception at the hands of deepfakes, or to preserve the authority of video footage. Moreover, the success of such (...) technologies depends on institutional trust that is in short supply. Finally, outsourcing the discrimination between the real and the fake to automated, largely opaque systems runs the risk of undermining epistemic autonomy. (shrink)

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)

The development and deployment of machine learning and artificial intelligence (AI) engender “AI colonialism”, a term that conceptually overlaps with “data colonialism”, as a form of injustice. AI colonialism is in need of decolonization for three reasons. Politically, because it enforces digital capitalism’s hegemony. Ecologically, as it negatively impacts the environment and intensifies the extraction of natural resources and consumption of energy. Epistemically, since the social systems within which AI is embedded reinforce Western universalism by imposing Western colonial values on (...) the global South when these manifest in the digital realm is a form of digital capitalism. These reasons require a new conceptualization of AI decolonization. First this paper draws from the historical debates on the concepts of colonialism and decolonization. Secondly it retrieves Achille Mbembe’s notion of decolonization as disenclosure to argue that the decolonization of AI will have to be the abolishment of political, ecological and epistemic borders erected and reinforced in the phases of its design, production, development of AI in the West and drawing from the knowledge from the global South. In conclusion, it is discussed how conceiving of decolonial AI as form of disenclosure opens up new ways to think about and intervene in colonial instantiations of AI development and deployment, in order to empower “the wretched of AI”, re-ecologise the unsustainable ecologies AI depends on and to counter the colonial power structures unreflective AI deployment risks to reinforce. (shrink)

AI-supported methods for identifying and combating disinformation are progressing in their development and application. However, these methods face a litany of epistemic and ethical challenges. These include (1) robustly defining disinformation, (2) reliably classifying data according to this definition, and (3) navigating ethical risks in the deployment of countermeasures, which involve a mixture of harms and benefits. This paper seeks to expose and offer preliminary analysis of these challenges.

The philosophy of AI has seen some changes, in particular: 1) AI moves away from cognitive science, and 2) the long term risks of AI now appear to be a worthy concern. In this context, the classical central concerns – such as the relation of cognition and computation, embodiment, intelligence & rationality, and information – will regain urgency.

Ikigai is a Japanese concept, which, in brief, refers to the “reason or purpose to live”. I-risks have been identified as a category of risks complementing x- risks, i.e., existential risks, and s-risks, i.e., suffering risks, which describes undesirable future scenarios in which humans are deprived of the pursuit of their individual ikigai. While some developments in AI increase i-risks, there are also AI-driven virtual opportunities, which reduce i-risks by increasing the space of potential ikigais, largely due to developments in (...) generative AI, virtual worlds as well as AI-driven hyper-personalization. The purpose of this paper is to present a first attempt to map the potential AI-driven virtual hyper-personalised ikigai universe. Moreover, challenges and further ideas are presented. (shrink)

The pervasiveness of AI-empowered technologies across multiple sectors has led to drastic changes concerning traditional social practices and how we relate to one another. Moreover, market-driven Big Tech corporations are now entering public domains, and concerns have been raised that they may even influence public agenda and research. Therefore, this chapter focuses on assessing and evaluating what kind of business model is desirable to incentivise the AI for Social Good (AI4SG) factors. In particular, the chapter explores the implications of this (...) discourse for SDG #17 (global partnership) and how this goal may encourage Big Tech corporations to strengthen multi-stakeholder partnerships that promote effective public-private and civil society partnerships and the meaningful co-presence of non-market and market values. In doing so, the chapter proposes an analysis of the sociological notion of "social license to operate" (SLO) elaborated in the mining and extractive industry literature and introduces it into the discourse on sustainable digital business models and responsible management of risks in the digital age. This serves to explore how such a social license can be adopted as a practice by digital business models to foster trust, collaboration and coordination among different actors - AI researchers and initiatives, institutions and civil society at large - for the support of SDGs interrelated targets and goals. (shrink)

Better instruments to predict the future evolution of artificial intelligence (AI) are needed, as the destiny of our civilization depends on it. One of the ways to such prediction is the analysis of the convergent drives of any future AI, started by Omohundro. We show that one of the convergent drives of AI is a militarization drive, arising from AI’s need to wage a war against its potential rivals by either physical or software means, or to increase its bargaining power. (...) This militarization trend increases global catastrophic risk or even existential risk during AI takeoff, which includes the use of nuclear weapons against rival AIs, blackmail by the threat of creating a global catastrophe, and the consequences of a war between two AIs. As a result, even benevolent AI may evolve into potentially dangerous military AI. The type and intensity of militarization drive depend on the relative speed of the AI takeoff and the number of potential rivals. We show that AI militarization drive and evolution of national defense will merge, as a superintelligence created in the defense environment will have quicker takeoff speeds, but a distorted value system. We conclude with peaceful alternatives. (shrink)

This is the editorial for a special volume of JETAI, featuring papers by Omohundro, Armstrong/Sotala/O’Heigeartaigh, T Goertzel, Brundage, Yampolskiy, B. Goertzel, Potapov/Rodinov, Kornai and Sandberg. - If the general intelligence of artificial systems were to surpass that of humans significantly, this would constitute a significant risk for humanity – so even if we estimate the probability of this event to be fairly low, it is necessary to think about it now. We need to estimate what progress we can expect, (...) what the impact of superintelligent machines might be, how we might design safe and controllable systems, and whether there are directions of research that should best be avoided or strengthened. (shrink)

On 8th August 2019, Secretary of State for Health and Social Care, Matt Hancock, announced the creation of a £250 million NHS AI Lab. This significant investment is justified on the belief that transforming the UK’s National Health Service (NHS) into a more informationally mature and heterogeneous organisation, reliant on data-based and algorithmically-driven interactions, will offer significant benefit to patients, clinicians, and the overall system. These opportunities are realistic and should not be wasted. However, they may be missed (one may (...) recall the troubled Care.data programme) if the ethical challenges posed by this transformation are not carefully considered from the start, and then addressed thoroughly, systematically, and in a socially participatory way. To deal with this serious risk, the NHS AI Lab should create an Ethics Advisory Board and monitor, analyse, and address the normative and overarching ethical issues that arise at the individual, interpersonal, group, institutional and societal levels in AI for healthcare. (shrink)

Where artificial agents are not liable to be ascribed true moral agency and responsibility in their own right, we can understand them as acting as proxies for human agents, as making decisions on their behalf. What I call the ‘Moral Proxy Problem’ arises because it is often not clear for whom a specific artificial agent is acting as a moral proxy. In particular, we need to decide whether artificial agents should be acting as proxies for low-level agents — e.g. individual (...) users of the artificial agents — or whether they should be moral proxies for high-level agents — e.g. designers, distributors or regulators, that is, those who can potentially control the choice behaviour of many artificial agents at once. Who we think an artificial agent is a moral proxy for determines from which agential perspective the choice problems artificial agents will be faced with should be framed: should we frame them like the individual choice scenarios previously faced by individual human agents? Or should we, rather, consider the expected aggregate effects of the many choices made by all the artificial agents of a particular type all at once? This paper looks at how artificial agents should be designed to make risky choices, and argues that the question of risky choice by artificial agents shows the moral proxy problem to be both practically relevant and difficult. (shrink)

That AI will have a major impact on society is no longer in question. Current debate turns instead on how far this impact will be positive or negative, for whom, in which ways, in which places, and on what timescale. In order to frame these questions in a more substantive way, in this prolegomena we introduce what we consider the four core opportunities for society offered by the use of AI, four associated risks which could emerge from its overuse or (...) misuse, and the opportunity costs associated with its under use. We then offer a high-level view of the emerging advantages for organisations of taking an ethical approach to developing and deploying AI. Finally, we introduce a set of five principles which should guide the development and deployment of AI technologies. The development of laws, policies and best practices for seizing the opportunities and minimizing the risks posed by AI technologies would benefit from building on ethical frameworks such as the one offered here. (shrink)

Engineering an artificial intelligence to play an advisory role in morally charged decision making will inevitably introduce meta-ethical positions into the design. Some of these positions, by informing the design and operation of the AI, will introduce risks. This paper offers an analysis of these potential risks along the realism/anti-realism dimension in metaethics and reveals that realism poses greater risks, but, on the other hand, anti-realism undermines the motivation for engineering a moral AI in the first place.

Failing to recognise sentience in AI systems (false negatives) poses a far greater risk of potentially astronomical suffering compared to mistakenly attributing sentience to non-sentient systems (false positives). This paper analyses the issue through the moral frameworks of longtermism, utilitarianism, and deontology, concluding that all three assign greater urgency to avoiding false negatives. Given the astronomical number of AIs that may exist in the future, even a small chance of overlooking sentience is an unacceptable risk. To address this, (...) the paper proposes a comprehensive approach including research, field-building, and tentative policy development. Humanity must take steps to ensure the well-being of all sentient minds, both biological and artificial. (shrink)

Alan Turing, one of the fathers of computing, warned that Artificial Intelligence (AI) could one day pose an existential risk to humanity. Today, recent advancements in the field AI have been accompanied by a renewed set of existential warnings. But what exactly constitutes an existential risk? And how exactly does AI pose such a threat? In this chapter we aim to answer these questions. In particular, we will critically explore three commonly cited reasons for thinking that AI poses (...) an existential threat to humanity: the control problem, the possibility of global disruption from an AI race dynamic, and the weaponization of AI. (shrink)

The main aim of this article is to reflect on the impact of biases related to artificial intelligence (AI) systems developed to tackle issues arising from the COVID-19 pandemic, with special focus on those developed for triage and risk prediction. A secondary aim is to review assessment tools that have been developed to prevent biases in AI systems. In addition, we provide a conceptual clarification for some terms related to biases in this particular context. We focus mainly on nonracial (...) biases that may be less considered when addressing biases in AI systems in the existing literature. In the manuscript, we found that the existence of bias in AI systems used for COVID-19 can result in algorithmic justice and that the legal frameworks and strategies developed to prevent the apparition of bias have failed to adequately consider social determinants of health. Finally, we make some recommendations on how to include more diverse professional profiles in order to develop AI systems that increase the epistemic diversity needed to tackle AI biases during the COVID-19 pandemic and beyond. (shrink)

Following the publication of numerous ethical principles and guidelines, the concept of 'Trustworthy AI' has become widely used. However, several AI ethicists argue against using this concept, often backing their arguments with decades of conceptual analyses made by scholars who studied the concept of trust. In this paper, I describe the historical-philosophical roots of their objection and the premise that trust entails a human quality that technologies lack. Then, I review existing criticisms about 'Trustworthy AI' and the consequence of ignoring (...) these criticisms: if the concept of 'Trustworthy AI' is kept being used, we risk attributing responsibilities to agents who cannot be held responsible, and consequently, deteriorate social structures which regard accountability and liability. Nevertheless, despite suggestions to shift the paradigm from 'Trustworthy AI' to 'Reliable AI', I argue that, realistically, this concept will be kept being used. I end by arguing that, ultimately, AI ethics is also about power, social justice, and scholarly activism. Therefore, I propose that community-driven and social justice-oriented ethicists of AI and trust scholars further focus on (a) democratic aspects of trust formation; and (b) draw attention to critical social aspects highlighted by phenomena of distrust. This way, it will be possible to further reveal shifts in power relations, challenge unfair status quos, and suggest meaningful ways to keep the interests of citizens. (shrink)

Artificial intelligence (AI) is becoming increasingly influential in most people’s lives. This raises many philosophical questions. One is what responsibility we have as individuals to guide the development of AI in a desirable direction. More specifically, how should this responsibility be distributed among individuals and between individuals and other actors? We investigate this question from the perspectives of five principles of distribution that dominate the discussion about responsibility in connection with climate change: effectiveness, equality, desert, need, and ability. Since much (...) is already written about these distributions in that context, we believe much can be gained if we can make use of this discussion also in connection with AI. Our most important findings are: (1) Different principles give different answers depending on how they are interpreted but, in many cases, different interpretations and different principles agree and even strengthen each other. If for instance ‘equality-based distribution’ is interpreted in a consequentialist sense, effectiveness, and through it, ability, will play important roles in the actual distributions, but so will an equal distribution as such, since we foresee that an increased responsibility of underrepresented groups will make the risks and benefits of AI more equally distributed. The corresponding reasoning is true for need-based distribution. (2) If we acknowledge that someone has a certain responsibility, we also have to acknowledge a corresponding degree of influence for that someone over the matter in question. (3) Independently of which distribution principle we prefer, ability cannot be dismissed. Ability is not fixed, however and if one of the other distributions is morally required, we are also morally required to increase the ability of those less able to take on the required responsibility. (shrink)

Concerns about threats to human autonomy feature prominently in the field of AI ethics. One aspect of this concern relates to the use of AI systems for problematically manipulative influence. In response to this, the European Union’s draft AI Act (AIA) includes a prohibition on AI systems deploying subliminal techniques that alter people’s behavior in ways that are reasonably likely to cause harm (Article 5(1)(a)). Critics have argued that the term ‘subliminal techniques’ is too narrow to capture the target cases (...) of AI-based manipulation. We propose a definition of ‘subliminal techniques’ that (a) is grounded on a plausible interpretation of the legal text; (b) addresses all or most of the underlying ethical concerns motivating the prohibition; (c) is defensible from a scientific and philosophical perspective; and (d) does not over-reach in ways that impose excessive administrative and regulatory burdens. The definition is meant to provide guidance for design teams seeking to pursue responsible and ethically aligned AI innovation. (shrink)

This paper aims to provide a roadmap to AI governance. In contrast to the reigning paradigms, we argue that AI governance should not be merely a reactive, punitive, status-quo-defending enterprise, but rather the expression of an expansive, proactive vision for technology—to advance human flourishing. Advancing human flourishing in turn requires democratic/political stability and economic empowerment. Our overarching point is that answering questions of how we should govern this emerging technology is a chance not merely to categorize and manage narrow (...) class='Hi'>risk but also to construe the risks and opportunities much more broadly, and to make correspondingly large investments in public goods, personnel, and democracy itself. To lay out this vision, we take four steps. First, we define some central concepts in the field, disambiguating between forms of technological harms and risks. Second, we review normative frameworks governing emerging technology that are currently in use around the globe. Third, we outline an alternative normative framework based in power-sharing liberalism. Fourth, we walk through a series of governance tasks that ought to be accomplished by any policy framework guided by our model of power-sharing liberalism. We follow these with proposals for implementation vehicles. (shrink)

Australia is a leading AI nation with strong allies and partnerships. Australia has prioritised the development of robotics, AI, and autonomous systems to develop sovereign capability for the military. Australia commits to Article 36 reviews of all new means and method of warfare to ensure weapons and weapons systems are operated within acceptable systems of control. Additionally, Australia has undergone significant reviews of the risks of AI to human rights and within intelligence organisations and has committed to producing ethics guidelines (...) and frameworks in Security and Defence. Australia is committed to OECD’s values-based principles for the responsible stewardship of trustworthy AI as well as adopting a set of National AI ethics principles. While Australia has not adopted an AI governance framework specifically for Defence; Defence Science has published ‘A Method for Ethical AI in Defence’ (MEAID) technical report which includes a framework and pragmatic tools for managing ethical and legal risks for military applications of AI. (shrink)

Initiatives relying on artificial intelligence (AI) to deliver socially beneficial outcomes—AI for social good (AI4SG)—are on the rise. However, existing attempts to understand and foster AI4SG initiatives have so far been limited by the lack of normative analyses and a shortage of empirical evidence. In this Perspective, we address these limitations by providing a definition of AI4SG and by advocating the use of the United Nations’ Sustainable Development Goals (SDGs) as a benchmark for tracing the scope and spread of AI4SG. (...) We introduce a database of AI4SG projects gathered using this benchmark, and discuss several key insights, including the extent to which different SDGs are being addressed. This analysis makes possible the identification of pressing problems that, if left unaddressed, risk hampering the effectiveness of AI4SG initiatives. (shrink)

We cooperate with other people despite the risk of being exploited or hurt. If future artificial intelligence (AI) systems are benevolent and cooperative toward us, what will we do in return? Here we show that our cooperative dispositions are weaker when we interact with AI. In nine experiments, humans interacted with either another human or an AI agent in four classic social dilemma economic games and a newly designed game of Reciprocity that we introduce here. Contrary to the hypothesis (...) that people mistrust algorithms, participants trusted their AI partners to be as cooperative as humans. However, they did not return AI's benevolence as much and exploited the AI more than humans. These findings warn that future self-driving cars or co-working robots, whose success depends on humans' returning their cooperativeness, run the risk of being exploited. This vulnerability calls not just for smarter machines but also better human-centered policies. (shrink)

Abstract: As there are no currently obvious ways to create safe self-improving superintelligence, but its emergence is looming, we probably need temporary ways to prevent its creation. The only way to prevent it is to create a special type of AI that is able to control and monitor the entire world. The idea has been suggested by Goertzel in the form of an AI Nanny, but his Nanny is still superintelligent, and is not easy to control. We explore here ways (...) to create the safest and simplest form of AI which may work as an AI Nanny, that is, a global surveillance state powered by a Narrow AI, or AI Police. A similar but more limited system has already been implemented in China for the prevention of ordinary crime. AI police will be able to predict the actions of and stop potential terrorists and bad actors in advance. Implementation of such AI police will probably consist of two steps: first, a strategic decisive advantage via Narrow AI created by an intelligence services of a nuclear superpower, and then ubiquitous control over potentially dangerous agents which could create unauthorized artificial general intelligence which could evolve into Superintelligence. (shrink)

Autonomy is often considered a core value of Western society that is deeply entrenched in moral, legal, and political practices. The development and deployment of artificial intelligence (AI) systems to perform a wide variety of tasks has raised new questions about how AI may affect human autonomy. Numerous guidelines on the responsible development of AI now emphasise the need for human autonomy to be protected. In some cases, this need is linked to the emergence of increasingly ‘autonomous’ AI systems that (...) can perform tasks without human control or supervision. Do such ‘autonomous’ systems pose a risk to our own human autonomy? In this article, I address the question of a trade-off between human autonomy and system ‘autonomy’. (shrink)

The debate about the ethical implications of Artificial Intelligence dates from the 1960s :741–742, 1960; Wiener in Cybernetics: or control and communication in the animal and the machine, MIT Press, New York, 1961). However, in recent years symbolic AI has been complemented and sometimes replaced by Neural Networks and Machine Learning techniques. This has vastly increased its potential utility and impact on society, with the consequence that the ethical debate has gone mainstream. Such a debate has primarily focused on principles—the (...) ‘what’ of AI ethics —rather than on practices, the ‘how.’ Awareness of the potential issues is increasing at a fast rate, but the AI community’s ability to take action to mitigate the associated risks is still at its infancy. Our intention in presenting this research is to contribute to closing the gap between principles and practices by constructing a typology that may help practically-minded developers apply ethics at each stage of the Machine Learning development pipeline, and to signal to researchers where further work is needed. The focus is exclusively on Machine Learning, but it is hoped that the results of this research may be easily applicable to other branches of AI. The article outlines the research method for creating this typology, the initial findings, and provides a summary of future research needs. (shrink)

From music recommendation to assessment of asylum applications, machine-learning algorithms play a fundamental role in our lives. Naturally, the rise of AI implementation strategies has brought to public attention the ethical risks involved. However, the dominant anti-discrimination discourse, too often preoccupied with identifying particular instances of harmful AIs, has yet to bring clearly into focus the more structural roots of AI-based injustice. This paper addresses the problem of AI-based injustice from a distinctively epistemic angle. More precisely, I argue that the (...) injustice generated by the implementation of AI machines in our societies is, in some paradigmatic cases, also a form of epistemic injustice. With a particular focus on AIs employed as gatekeepers of our epistemic resources, this paper shows how their epistemically conformist behaviour is responsible for the marginalisation and the ostracism of minoritarian perspectives. Because it clarifies key structural flaws and weaknesses of current AI design, this paper helps make headway in critical discussion of current AI technologies. And because it forges new theoretical tools to understand forms of epistemic oppression, this paper also contributes to the advancement of feminist theorisation. (shrink)

Abstract: As there are no visible ways to create safe self-improving superintelligence, but it is looming, we probably need temporary ways to prevent its creation. The only way to prevent it, is to create special AI, which is able to control and monitor all places in the world. The idea has been suggested by Goertzel in form of AI Nanny, but his Nanny is still superintelligent and not easy to control, as was shown by Bensinger at al. We explore here (...) the ways to create the safest and simplest form of AI, which may work as AI Nanny. Such AI system will be enough to solve most problems, which we expect the AI will solve, including control of robotics, acceleration of the medical research, but will present less risk, as it will be less different from humans. As AI police, it will work as operation system for most computers, producing world surveillance system, which will be able to envision and stop any potential terrorists and bad actors in advance. As uploading technology is lagging, and neuromorphic AI is intrinsically dangerous, the most plausible way to human-based AI Nanny is either functional model of the human mind or a Narrow-AI empowered group of people. (shrink)

Discussions about the possible consequences of creating superintelligence have included the possibility of existential risk, often understood mainly as the risk of human extinction. We argue that suffering risks (s-risks) , where an adverse outcome would bring about severe suffering on an astronomical scale, are risks of a comparable severity and probability as risks of extinction. Preventing them is the common interest of many different value systems. Furthermore, we argue that in the same way as superintelligent AI both (...) contributes to existential risk but can also help prevent it, superintelligent AI can both be a suffering risk or help avoid it. Some types of work aimed at making superintelligent AI safe will also help prevent suffering risks, and there may also be a class of safeguards for AI that helps specifically against s-risks. (shrink)

Humans and artificial intelligences (AI) will increasingly participate digitally and physically in conflicts yet there is a lack of trusted communications across agents and platforms. For example, humans in disasters and conflict already use messaging and social media to share information, however, international humanitarian relief organisations treat this information as unverifiable and untrustworthy. AI may reduce the ‘fog-of-war’ and improve outcomes, however current AI implementations are often brittle, have a narrow scope of application and wide ethical risks. Meanwhile, human error (...) causes significant civilian harms even by combatants committed to complying with international humanitarian law. AI offers an opportunity to help reduce the tragedy of war and better deliver humanitarian aid to those who need it. However, to be successful, these systems must be trusted by humans and their information systems, overcoming flawed information flows in conflict and disaster zones that continue to be marked by intermittent communications, poor situation awareness, mistrust and human errors. In this paper, we consider the integration of a communications protocol (the ‘Whiteflag protocol’), distributed ledger technology, and information fusion with artificial intelligence (AI), to improve conflict communications called “Protected Assurance Understanding Situation & Entities” PAUSE. Such a trusted human-AI communication network could provide accountable information exchange regarding protected entities, critical infrastructure; humanitarian signals and status updates for humans and machines in conflicts. Trust-based information fusion provides resource-efficient use of diverse data sources to increase the reliability of reports. AI can be used to catch human mistakes and complement human decision making, while human judgment can direct and override AI recommendations. We examine several case studies for the integration of these technologies into a trusted human-AI network for humanitarian benefit including mapping a conflict zone with civilians and combatants in real time, preparation to avoid incidents and using the network to manage misinformation. (shrink)

What does Artificial Intelligence (AI) have to contribute to health care? And what should we be looking out for if we are worried about its risks? In this paper we offer a survey, and initial evaluation, of hopes and fears about the applications of artificial intelligence in medicine. AI clearly has enormous potential as a research tool, in genomics and public health especially, as well as a diagnostic aid. It’s also highly likely to impact on the organisational and business practices (...) of healthcare systems in ways that are perhaps under-appreciated. Enthusiasts for AI have held out the prospect that it will free physicians up to spend more time attending to what really matters to them and their patients. We will argue that this claim depends upon implausible assumptions about the institutional and economic imperatives operating in contemporary healthcare settings. We will also highlight important concerns about privacy, surveillance, and bias in big data, as well as the risks of over trust in machines, the challenges of transparency, the deskilling of healthcare practitioners, the way AI reframes healthcare, and the implications of AI for the distribution of power in healthcare institutions. We will suggest that two questions, in particular, are deserving of further attention from philosophers and bioethicists. What does care look like when one is dealing with data as much as people? And, what weight should we give to the advice of machines in our own deliberations about medical decisions? (shrink)