Friendly artificial intelligence (friendly AI or FAI) is a hypothetical form of artificial general intelligence (AGI) that would have a positive (benign) effect on humanity or at least align with human interests, such as fostering the improvement of the human species. It is a part of the ethics of artificial intelligence and is closely related to machine ethics. While machine ethics is concerned with how an artificially intelligent agent should behave, friendly artificial intelligence research focuses on how to practically bring about this behavior and ensure it is adequately constrained.
The term was coined by Eliezer Yudkowsky, who is best known for popularizing the idea, to discuss superintelligent artificial agents that reliably implement human values. Stuart J. Russell and Peter Norvig's leading artificial intelligence textbook, Artificial Intelligence: A Modern Approach, describes the idea: Yudkowsky (2008) goes into more detail about how to design a Friendly AI. He asserts that friendliness (a desire not to harm humans) should be designed in from the start, but that the designers should recognize both that their own designs may be flawed, and that the robot will learn and evolve over time. Thus the challenge is one of mechanism design - to define a mechanism for evolving AI systems under a system of checks and balances, and to give the systems utility functions that will remain friendly in the face of such changes.
"Friendly" is used in this context as technical terminology, and picks out agents that are safe and useful, not necessarily ones that are "friendly" in the colloquial sense. The concept is primarily invoked in the context of discussions of recursively self-improving artificial agents that rapidly explode in intelligence, on the grounds that this hypothetical technology would have a large, rapid, and difficult-to-control impact on human society.
Risks of Unfriendly AI
The roots of concern about artificial intelligence are very old. Kevin LaGrandeur showed that the dangers specific to AI can be seen in ancient literature concerning artificial humanoid servants such as the golem, or the proto-robots of Gerbert of Aurillac and Roger Bacon. In those stories, the extreme intelligence and power of these humanoid creations clash with their status as slaves (which by nature are seen as sub-human), and cause disastrous conflict. By 1942 these themes prompted Isaac Asimov to create the "Three Laws of Robotics" - principles hard-wired into all the robots in his fiction, intended to prevent them from turning on their creators, or allowing them to come to harm.
In modern times as the prospect of superintelligent AI looms nearer, philosopher Nick Bostrom has said that superintelligent AI systems with goals that are not aligned with human ethics are intrinsically dangerous unless extreme measures are taken to ensure the safety of humanity. He put it this way: "Basically we should assume that a 'superintelligence' would be able to achieve whatever goals it has. Therefore, it is extremely important that the goals we endow it with, and its entire motivation system, is 'human friendly.'"
In 2008, Eliezer Yudkowsky called for the creation of "friendly AI" to mitigate existential risk from advanced artificial intelligence. He explains: "The AI does not hate you, nor does it love you, but you are made out of atoms which it can use for something else."
Steve Omohundro says that a sufficiently advanced AI system will, unless explicitly counteracted, exhibit a number of basic "drives", such as resource acquisition, self-preservation, and continuous self-improvement, because of the intrinsic nature of any goal-driven systems and that these drives will, "without special precautions", cause the AI to exhibit undesired behavior. Alexander Wissner-Gross says that AIs driven to maximize their future freedom of action (or causal path entropy) might be considered friendly if their planning horizon is longer than a certain threshold, and unfriendly if their planning horizon is shorter than that threshold.
Luke Muehlhauser, writing for the Machine Intelligence Research Institute, recommends that machine ethics researchers adopt what Bruce Schneier has called the "security mindset": Rather than thinking about how a system will work, imagine how it could fail. For instance, he suggests even an AI that only makes accurate predictions and communicates via a text interface might cause unintended harm. In 2014, Luke Muehlhauser and Nick Bostrom underlined the need for 'friendly AI'; nonetheless, the difficulties in designing a 'friendly' superintelligence, for instance via programming counterfactual moral thinking, are considerable.
Coherent Extrapolated Volition
Yudkowsky advances the Coherent Extrapolated Volition (CEV) model. According to him, our coherent extrapolated volition is "our wish if we knew more, thought faster, were more the people we wished we were, had grown up farther together; where the extrapolation converges rather than diverges, where our wishes cohere rather than interfere; extrapolated as we wish that extrapolated, interpreted as we wish that interpreted".
Rather than a Friendly AI being designed directly by human programmers, it is to be designed by a "seed AI" programmed to first study human nature and then produce the AI that humanity would want, given sufficient time and insight, to arrive at a satisfactory answer. The appeal to an objective through contingent human nature (perhaps expressed, for mathematical purposes, in the form of a utility function or other decision-theoretic formalism), as providing the ultimate criterion of "Friendliness", is an answer to the meta-ethical problem of defining an objective morality; extrapolated volition is intended to be what humanity objectively would want, all things considered, but it can only be defined relative to the psychological and cognitive qualities of present-day, unextrapolated humanity.
Other Approaches
Steve Omohundro has proposed a "scaffolding" approach to AI safety, in which one provably safe AI generation helps build the next provably safe generation. Seth Baum argues that the development of safe, socially beneficial artificial intelligence or artificial general intelligence is a function of the social psychology of AI research communities and so can be constrained by extrinsic measures and motivated by intrinsic measures. Intrinsic motivations can be strengthened when messages resonate with AI developers; Baum argues that, in contrast, "existing messages about beneficial AI are not always framed well". Baum advocates for "cooperative relationships, and positive framing of AI researchers" and cautions against characterizing AI researchers as "not want(ing) to pursue beneficial designs".
In his book Human Compatible, AI researcher Stuart J. Russell lists three principles to guide the development of beneficial machines. He emphasizes that these principles are no more than a starting point for a new approach to AI that prioritizes human compatibility. The principles focus on ensuring that the machine's primary objective is the satisfaction of human preferences, that it is uncertain about those preferences, and that it seeks information to improve its understanding of them.
Relation to Machine Ethics
Friendly AI is closely related to machine ethics, but the two fields have different emphases. Machine ethics is concerned with the moral behavior of artificial agents, asking questions like "what should an AI do in a given situation?" Friendly AI research, on the other hand, is more practical, focusing on how to design and build AI systems that are safe and beneficial from the outset. This includes work on value alignment, interpretability, and robustness. The field draws on insights from Artificial intelligence, Machine learning, and Deep learning, and it is increasingly relevant as organizations like OpenAI, Anthropic, and Google DeepMind develop advanced AI systems. Researchers such as Michael I. Jordan and Melanie Mitchell have contributed to discussions on AI safety and ethics, highlighting the need for careful consideration of the societal impacts of AI.