The Ethics of Artificial Intelligence and Machine Learning

Artificial Intelligence (AI) and machine learning are rapidly becoming integral parts of modern society. As we learn to coexist and even, in some cases, rely on these tools, we must also traverse the ethical issues associated with them.

One ethical issue that arises with the continued emergence of AI and machine learning concerns privacy and surveillance. The data collected on a user by consistent use of “free” – that is, systems that are paid for by external parties to the user – services on the internet creates a data trail that can be used by AI and machines. These details can be employed to manipulate search results and promote corporations, potentially in an unethical fashion. 

Another ethical concern manifests in what is unknown in the Artificial Intelligence (AI) systems. With machine learning being the basis of a good portion of AI systems, it’s fundamental that we understand what we both do and do not know about their functions. While the originating system and its parameters are created by a human supervisor, the process will eventually evolve to be automated. At this point, the full processing activity is unknown even to the creator of the AI.

This discussion highlights the issue of responsibility. The European Group on Ethics in Science and New Technologies (2018) upheld that accountability, liability, and the rule of law were basic requirements that applied to new technologies. The debate of when accountability transfers from the creator of the machine to the machine itself has sparked several new discussions.

Bias is an immediate ethical issue in many aspects of digital humanities, Artificial Intelligence (AI) and machine learning are no exception. Müller (2020) outlined three main forms of bias in AI: bias based on a cognitive feature of a person that becomes embedded in the machine system, cognitive bias based on human interpretation of information, and systematic bias.

The ethical implications of AI and machine learning also extend to and envelop areas such as employment. Digital automation presents an opportunity for easily duplicatable and consistently affordable options to replace human thought and cognitive processing tasks (Bostrom & Yudkowsky, 2014). The issue of enhanced levels of unemployment as a consequence of digital automation poses a previously untraversed ethical dilemma: what happens to the millions of workers whose jobs become irrelevant?

To conclude, Artificial Intelligence (AI) and machine learning now provide a fundamental basis for a considerable portion of modern society’s activities and systems. The ethical issues that arise due to these developments include privacy and surveillance monitoring, unknown functionalities, debates over responsibility, bias in systems, and issues of unemployment. The likelihood that modern society will continue to form a reliance on AI and machine learning makes understanding these ethical concerns even more instrumental.

References:

Bostrom, N., Yudkowsky, E. (2014). “The Ethics of Artificial Intelligence”. The Cambridge Handbook of Artificial Intelligence, Cambridge, 316-334. 

European Group on Ethics in Science and New Technologies (2018). “Statement on Artificial Intelligence, Robotics and ‘Autonomous’ Systems”, European Commission. 

Müller, V. C. (2020). “Ethics of Artificial Intelligence and Robotics”, The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University.

Comments
All comments.
Comments