Catalog of General Ethical Requirements for AI Certification - Workshop with Jocelyn Maclure and Markus Gabriel

"General Ethical Requirements for AI Certification"  a conversation with 
Jocelyn Maclure and Markus Gabriel and launch of the whitepaper by Nicholas 
Kluge Corrêa and Julia Maria Mönig.

The research project "Certified AI" explores the question of what a 
certification of "trustworthy AI" could look like. The philosophy team, Nicholas 
Kluge Corrêa and Julia Maria Mönig, research this question from an ethical 
perspective.

While many arguments can be raised against certifying "ethical" 
AI, with our latest publication we offer guidance to computer scientists and 
programmers on how to try to make AI-based applications more "ethical". We do 
this by presenting (a list of) ethical principles that are consistent with 
other proposals in the literature. We briefly explain what the values are and 
why they are important, and give examples to align them with the risk 
categories of the European AI Act (2024). To operationalize the principles, 
for each step we propose methods and technical tools that can be used to put 
the ethical requirements into practice. 
In the long run, this could serve as a way to make the tech industry's current 
profit-driven motto of "move fast and break things" fairer, more privacy-
preserving, safer and more robust, sustainable and transparent, while 
promoting accountability and truthfulness, if developers adhere to the 
proposed practices. 

Arguments against ethics 'certification' (including by the authors) are that 
companies and other stakeholders could use it as 'ethics washing'. In 
addition, practices may evolve over time and therefore certification (as in 
other areas) could only be valid for a limited period of time, if at all, 
after which it would need to be renewed. In addition, ethical tools may not 
have the intended effects, as has been shown in the literature, e.g. for the 
example of fairness algorithms. Last but not least, it is difficult to 
'standardise' ethics, as this entails quantification, whereas ethics refers to 
and consists of deliberation, with values that may change over time, and in a 
liberal-democratic system must take into account the views of all 
stakeholders, especially vulnerable groups. We hence suggest that the efforts 
laid out in the paper would need to be accompanied for instance by workshops, 
guidance by professional ethicists and discussions with involved stakeholders.

While the values presented in the white paper resonate with those usually 
attributed to the so-called "Global North", we understand these values as 
being universal, with human dignity and respect at their core. The four 
overarching principles of autonomy, beneficence, no harm and accountability are 
intended to serve all humans on earth, and call for the ethical inclusion of 
all potentially affected stakeholders,  e.g. through the principle of 
sustainability that addresses problems caused by the actions of the "Global 
North" on the "Global South".

In this workshop we will present the guidelines in the presence of Markus 
Gabriel, the principal investigator of the philosophy subproject of 
"Zertifizierte KI", and Jocelyn Maclure who will comment on them from a 
philosophical perspective, drawing on his expertise and decade-long experience 
in AI ethics.

More about the Research Project "Certified AI"