Artificial Morality: Top-down, Bottom-up, and Hybrid Approaches

By C. Allen, I. Šmit, Wendell Wallach. 2005

334 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Information Systems and Management, Safety Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Information Systems and Management

In Safety Research

Citations in the corpus

Value Alignment or Misalignment - What Will Keep Systems Accountable? (2017, background): “On the performance side, there have been ongoing questions, but few spelled-out programs, for what evaluation or testing of an autonomous system’s ethics could look like (e.g., the idea of a “moral Turing Test” (Allen, Smit, and Wallach 2005; Arnold and Scheutz 2016)).”
Ethics of artificial intelligence and robotics (2020, background): “The earlier authors (Allen, Smit, and Wallach 2005; Moor 2006) propose several ways to achieve explicit ethical agents, via programming it in (operational morality), via ‘developing’ the ethics itself (functional morality) and finally full-blown morality with full intelligence and sentience.”
The Challenge of Value Alignment: from Fairer Algorithms to AI Safety (2021, background): “The relevant ethical principles could derive from religious ideals, moral codes, culturally endorsed values, or philosophical systems (Allen et al 2005, 150).”
The Ethical Implications of Using Artificial Intelligence in Auditing (2020, background): “With this backdrop, ethics researchers have sought the best approach for identifying the ethical risks surrounding emerging technology during their introductory phase (Allen et al. 2005; Wright 2011; Brey 2012; Stahl et al. 2017).”
Six Human-Centered Artificial Intelligence Grand Challenges (2023, neutral): “However, it lacks a concept of human values, common sense, or ethics (Allen et al., 2005; Han et al., 2022), and thus may perform its tasks or influence decisions from a basis that is not human-oriented or may cause direct harms.”
Symbolic Behaviour in Artificial Intelligence (2021, background): “[61] Colin Allen, Iva Smit, and Wendell Wallach.”
Ethical by Design: Ethics Best Practices for Natural Language Processing (2017, background): “There is a body of work within philosophy on information ethics (Allen et al., 2005; Bynum, 2008); big data has created its own challenges, which are begin-2 ning to be discussed.”
Social choice ethics in artificial intelligence (2017, background): “An example of this appears above in the quote ‘‘[Bottom–up ethics] attempts to train or evolve agents whose behavior emulates morally praiseworthy human behavior’’ (Allen et al. 2005: 149; emphasis added).”
Interpretable Multi-Objective Reinforcement Learning through Policy Orchestration (2018, methodology): “A popular technique is called the bottom-up approach, i.e., teaching a machine what is right and wrong by example (Allen, Smit, and Wallach 2005).”