Anna Katharina Boos

This article advances a novel approach to the question of who should take responsibility when no one is to blame, and shows how it can be applied to the emerging problem of faultless harm caused by artificial intelligence (AI). As AI systems increase in complexity and scale, so too does the risk of harm that cannot be neatly traced back to human intent or negligence. This trajectory challenges conventional, blame-based responsibility practices, calling instead for an account that does not presuppose wrongdoing. I argue here that AI deployers bear a strict moral duty to take responsibility for harmful outcomes, even when the deployment is, all things considered, morally justified. This duty arises from what I call the ‘accidental relationship’, a distinctive moral relationship that emerges whenever an agent, here the AI deployer, non-negligently and non-intentionally inflicts harm on another party. Through this distinct agential involvement in the harm, the deployer incurs special duties to acknowledge and redress the victim’s loss. The present account does not conceptualize the duty to take responsibility as a matter of repairing a prior wrongdoing, but as a matter of affirming the victim’s equal moral standing within the accidental relationship.

This article advances a novel approach to the question of who should take responsibility when no one is to blame, and shows how it can be applied to the emerging problem of faultless harm caused by artificial intelligence (AI). As AI systems increase in complexity and scale, so too does the risk of harm that cannot be neatly traced back to human intent or negligence. This trajectory challenges conventional, blame-based responsibility practices, calling instead for an account that does not presuppose wrongdoing. I argue here that AI deployers bear a strict moral duty to take responsibility for harmful outcomes, even when the deployment is, all things considered, morally justified. This duty arises from what I call the ‘accidental relationship’, a distinctive moral relationship that emerges whenever an agent, here the AI deployer, non-negligently and non-intentionally inflicts harm on another party. Through this distinct agential involvement in the harm, the deployer incurs special duties to acknowledge and redress the victim’s loss. The present account does not conceptualize the duty to take responsibility as a matter of repairing a prior wrongdoing, but as a matter of affirming the victim’s equal moral standing within the accidental relationship.

Despite growing interest in automated (or algorithmic) decision-making (ADM), little work has been done to conceptually clarify the term. This article aims to tackle this issue by developing a conceptualization of ADM specifically tailored to organizational contexts. It has two main goals: (1) to meaningfully demarcate ADM from similar, yet distinct algorithm-supported practices; and (2) to draw internal distinctions such that different ADM types can be meaningfully distinguished. The proposed conceptualization builds on three arguments: First, ADM primarily refers to the automation of practical decisions (decisions to _φ_) as opposed to cognitive decisions (decisions that _p_). Second, rather than referring to algorithms as literally making decisions, ADM refers to the use of algorithms to solve decision problems at an organizational level. Third, since algorithmic tools by nature primarily settle cognitive decision problems, their classification as ADM depends on whether and to what extent an algorithmically generated output _p_ has an action triggering effect—i.e., translates into a consequential action _φ_. The examination of precisely this _p_-_φ_ relationship, allows us to pinpoint different ADM types (suggesting, offloading, superseding). Taking these three arguments into account, we arrive at the following definition: ADM refers to the practice of using algorithms to solve decision problems, where these algorithms can play a suggesting, offloading, or superseding role relative to humans, and decisions are defined as action triggering choices.