A Creed for AI and ML Practitioners
Several mentions of an ethical code of conduct or oath for Artificial Intelligence (AI) or Machine Learning (ML) practitioners have become headlines recently.
One of the causes is the discussion whether AI will be beneficial or the doom of humanity. Another cause are several instances in which ML has been used to it potential but also to a skewed and ethically questionable purpose. (That is a topic for an entire different post)
I fully support such an ethical oath or code of conduct, as no matter what one’s position in the long-term outlook of ML and AI are, such code ensures an ethical baseline. Sure, not everyone will be bound by such an oath, but adhering -and being able to reference- to such a code gives confidence into an individuals’ work.
Until a few days ago, I was not aware of an actual proposed code, and the discussions and the code I am aware of are lacking a in my opinion very important aspects: The effect any AI or ML system will have on the ethical and mental wellbeing as well as the long-term effects on our ecological and economic environment.
So, I am proposing this Machine Learning Practitioners Creed (MLPC). I used two well established and in the respective fields well known documents as a basis, combined and expanded on those: The first is the Physician’s Pledge published by the General Assembly of the World Medical Association (Declaration of Geneva), which is one form of the Hippocratic Oath. The second is the Aircraft Mechanic’s Creed (FAA).
I welcome comments, suggestions and a discussion.
Machine Learning Practitioners Creed
I dedicate my life primary to the service of all humanity and secondary to my works.
I shall make the health and wellbeing (bodily, mentally, ecologically and economically) of all entities my first consideration.
I will maintain utmost respect for human life, human societies and ecological dependencies; in the current moment as in futures to come.
I will not permit any consideration — neither explicit nor implicit (e.g. via biased data or assumption) — to intervene in the neutral discharge of my duty.
I shall not approve of work that I feel is beyond the limits of my skill or knowledge.
I shall not allow anyone to persuade me to approve of work against my better judgment.
I shall not permit my judgment to be influenced by money, other personal gain or thread.
I shall include in my work a note of caution if I feel it warranted or to prevent misuse. Such note is to be kept with my works always accessible and legible and can only be altered, amended or removed by me.
I will respect the secrets and trust confined in me, during and after conclusion of my work.
I will practice with conscience and dignity and in accordance with good scientific practice.
I will foster the honour, reliability, dependability and trustworthiness of ML, my profession and the ML community.
I will give to my teachers, colleagues, and students the respect and gratitude that is their due.
I will share my knowledge for the benefit of the human society, the advancement of scientific and furthering practical implementations of ML.
I will attend to my own health, well-being, and abilities in order to produce works of the highest standards.
I will not use my knowledge to infringe or violate human rights, civil liberties, societal norms and customs or applicable laws, even under threat.
I know that the autonomy, safety, privacy and dignity of each entity depends on my respect, skill and judgment: I shall never knowingly or against my judgment subject any entity to a risk or situation I would not assume for myself or for those dear to me.
I pledge to hold sacred the rights, privileges and possibilities conferred upon me as a ML practitioner. I make these promises solemnly, freely and upon my honour and in the knowledge that this places special obligations upon me.
About:
This Creed shall be understood and interpreted by its’ principles and ideas and NOT exclusively by its text and letters.
It shall be applicable for Machine Learning (ML) and AI Practitioners and encompasses their complete works and accomplishments (works), indiscriminate of the wider field.
It considers entities to be individuals (legal and natural), groups and societies — indiscriminate of age, disease or disability, creed, ethnic origin, gender, nationality, religion, political affiliation, race, sexual orientation, social standing, ecological or economic aspects, composition, location or any other factor.
It does not differentiate direct and indirect effects.
This post is intended to START a conversation rather than end one!
Thanks for reading!
Disclosure: This post reflects my personal opinion. It is not implied that this post reflects the opinion or stance of my employer or my employer’s customers.
Please keep the comments courteous and considerate of other people’s opinion and values.
