Ottawa Developing Artificial Intelligence Strategy for Public Service
Ottawa is working on a new artificial intelligence strategy for federal government operations, according to Treasury Board President Anita Anand on May 27.
“This is a significant step towards a unified and consistent federal approach to AI,” Ms. Anand stated during a meeting of AI experts in Gatineau, Que., where the strategy is under discussion.
The goal is to enhance government efficiency, improve services for Canadians, and strengthen scientific and research capabilities.
Anand emphasized the importance of streamlining interactions between the Canadian population, organizations, businesses, and federal services to achieve greater efficiencies.
While AI can automate routine tasks, Anand noted that generative AI will not be used for confidential matters such as those involving cabinet confidences.
She clarified that the initiative is not aimed at reducing jobs in the federal public service.
The initiative will involve establishing a specialized division to retrain current public servants. The government also plans to rapidly and competitively recruit top tech talent.
In addition to the expert meeting on Monday, the government will engage in broad consultations in the coming months before unveiling the strategy. There is no set deadline for its completion.
The federal government has already started integrating AI into its operations, utilizing the technology in numerous ways.
As the largest employer in Canada, the federal government serves as a model for private sector employers, according to Ms. Anand.
She emphasized the need for responsible AI usage, which includes respecting privacy laws and avoiding discriminatory or unethical purposes.
The announcement coincided with the signing of the voluntary federal code of conduct for AI by eight more private organizations, including Mastercard, Lenovo, and the Quebec real estate brokers’ association.
Participants in the code must commit to various measures aimed at minimizing AI risks, such as screening data sets for biases and monitoring systems for potential harms.
