This is part one of a series of blog articles about ethics and AI.
Blue J’s long-term vision imagines a world in which, thanks to advances in artificial intelligence and machine learning technology, legal uncertainty is a thing of the past. Recognizing that we must hold ourselves accountable to stay true to our mission and, in so doing, avoid creating unintended outcomes, Blue J’s internal ethics committee RELAI (which stands for Responsible, Ethical, Legal AI) was founded in 2019. Our first task was to draft a Charter on Algorithmic Responsibility.
In adopting this Charter, we hope to achieve two distinct aims.
First, we hope to codify and enforce the ways in which ethical decision making is built into every step of our product development processes. In developing our Charter, RELAI is attempting to capture and distil the aspects of Blue J’s current workflow that we want to preserve and enhance. This process is outlined in more detail in Part 2 of this blog post series.
These principles are not aspirational. They reflect what we currently do and plan to continue doing, even as we grow as a company and expand into new markets and new legal areas. By recording and sharing the ways in which ethical decision making is built into every aspect of our company, we are affirming that we will stay true to the values that have guided us since the beginning.
Second, we hope to be leaders in the nascent field of ethical AI. There is great interest from the public, and from all levels of government in the ethical use of artificial intelligence technology. It is expected that Charters of this type will soon become standard and ours can serve as a template and an example that other artificial intelligence companies can rely upon for guidance as to the best practices in the development of ethical artificial intelligence
In drafting our Charter, we were influenced by leading industry guidance on the application of ethical AI, as encapsulated by the Treasury Board Secretariat of Canada’s Directive on Automated Decision Making, and by the Montréal Declaration for a responsible development of artificial intelligence.
We then applied these industry leading standards to our own work, developing a set of guidelines that are unique to Blue J and can set an example to others. Once adopted by RELAI, the Charter was then reviewed and approved by key stakeholders, including senior Blue J’s senior leadership.