Despite its successes, to date, Artificial Intelligence (AI) is still characterized by a number of shortcomings with regard to different application domains and goals. These limitations are arguably both conceptual (e.g., related to the underlying theoretical models, such as symbolic vs.connectionist), and operational (e.g., related to robustness and ability to generalize). Biologically inspired AI, specifically brain-inspired AI, promises to provide further biological aspects beyond those already traditionally included in AI, making it possible to assess and possibly overcome some of its present shortcomings. 

A recent study examines some conceptual, technical, and ethical issues raised by the development and use of brain-inspired AI. Against this background, the paper asks whether there is anything ethically unique about brain-inspired AI. The aim of the paper is to introduce a method that has a heuristic nature and that can be applied to identify and address the ethical issues arising from brain-inspired AI.

Because intelligence is a biological phenomenon, Artificial Intelligence (AI) is biologically inspired in its very essence. Nevertheless, AI may mimic biology to varying degrees.In some cases, AI research might be directly and extensively inspired by biology in general and by the brain in particular. The increased collaboration between neuroscience and AI could result in further improvement of AI, possibly allowing it to overcome some of its current limitations. Biological inspiration is not limited to human brains and cognitive reasoning: it may also come from organisms, processes, and phenomena occurring at different spatial and temporal scales.

What is Brain-inspired AI?

In its broadest sense, biological inspiration refers to the compatibility of AI with current knowledge in biology, particularly in neurobiology. Such a general description, while useful for the sake of introducing the concept, is not sufficiently constrained to be technically operationalizable. More specifically, an AI system is biologically inspired when its architecture and functioning include biological constraints that make specific parts of the system biologically realistic. Importantly, a biologically inspired AI system does not necessarily fully emulate or replicate the reference biological system, since different levels of biological realism are possible. 

When compared to the human brain, current AI reveals a number of differences and limitations with regards to different domains and goals. These limitations are arguably of two main kinds: technical and conceptual. The technical limitations depend on the current technological stage of AI and are likely to be reduced and possibly overcome through further progress of knowledge and emerging technology. The conceptual limitations depend on the AI paradigms used, so overcoming them may require revised or new paradigms. To illustrate, present AI is still narrow, that is, it works for specific tasks in particular domains for which it is programmed and trained, and fails if environmental conditions are different from those in the training context.

Strategy for improvement

Brain-inspired AI is an attractive strategy for improving current AI, but it raises a number of technical and ethical issues. The study summarized the main conceptual and technical aspects of brain-inspired AI, and introduced a method for its ethical analysis, distinguishing two main kinds of ethical issues: practical and fundamental/foundational.

The researchers proposed that the practical issues arising from brain-inspired AI can be organized in terms of the following main levels:

• Operational, related to how AI works;
• Instrumental, related to how people use AI;
• Relational, related to how people see AI and to the resulting psychological and metaphysical human-AI relationship;
• Societal, related to the social and economic costs and consequences of the development and use of AI.

Fundamental/foundational ethical issues arising from brain-inspired AI as those issues that concern the justification of the attempt itself to build brain-inspired AI and its impact on how we think about fundamental moral notions.

According to the research brain-inspired AI has the potential to raise new fundamental/foundational and practical ethical issues, as well as to exacerbate the practical ethical issues raised by traditional AI, which should be considered in relation to this promising approach.

Full study.