Ethics can technically be applied to machines but the problem concerns the definition of a universally accepted ethical framework. The ethical construct as a human thing, subjected to cultural, religious, social norms, makes it such a problem to develop a single, universal means. However, fairness, transparency, accountability, harm prevention or other ethical principles can be embedded into machines decision making, as AI or autonomous systems. The problem lies in establishing what to apply ethics to: things that are ethically right in one culture are not necessarily right in another. For example, China’s state surveillance norms versus the norms of the EU (GDPR driven). According to the same type of disparity, for example, there is not one universal ethical view on lethal autonomous weapons, biased AI models, or labour automation. Therefore, AI ethics are often locally specific, in other words, given by local laws and values. However, despite these challenges, several people are attempting to come up with universal AI ethics in the form of the Asilomar AI Principles, the UNESCO’s Ethics of AI, and other industry led frameworks suggesting transparency and fairness. Contextual ethics, on the other hand, is a practical approach to ethics for AI, where it changes its ethical decision depending on the culture or law of the surrounding environment, standing as another international company would when having to comply with different legal frameworks from country to country. Usually its just trying to balance ethical universality with cultural specificity while ensuring ethical AI remains aligned with human values.

I think that the question of the existence of universal ethics for artificial intelligence is as utopian as the existence of a world government or a single religion. Reflections in this direction will inevitably lead to the creation of some kind of average monster, which will be equally inconvenient for all participants in communication, and its main and undoubted advantage will be the ability to control various human communities with greater ease. Thus, I believe that reflections on universal ethics for AI are anti-ethical, immoral. In any case, the ethics of machines will turn out to be either a copy of human ethics in all its diversity, or some kind of artificial construct, where the values ​​of artificial intelligence will be put at the forefront, and human beings will be in a subordinate position. Agree, such a prospect does not seem very bright to us.

Aksana Zaretskaya yes, I also agree with you regarding the difficulty of achieving this dilemma of creating a universal ethics to machines that aligns with the goals and desires of different societies. Also, creating this universal moral machine may be achievable and appropriate for some societies but not others, and this is the most that machines or artificial intelligence in general will achieve. Thank you.

I agree with you about the difficulty of applying this globally, due to cultural, religious, and value differences and variations, which makes unifying these ethical principles impossible to achieve and implement. Thank you for your comment. Prinz Jewel Ipunla Marquez