New Canadian Institute to advance AI security research
“In a fast-changing world full of uncertainties, Canadians can be confident that we will always take the necessary steps to ensure AI is safe, responsible, and reliable.”
The Canadian government announced on Tuesday that the Canadian Institute for Advanced Research (CIFAR) will collaborate with the Canadian Artificial Intelligence Safety Institute (CAISI) to advance AI safety initiatives.
This new institute is a significant step toward ensuring the safe development and implementation of artificial intelligence (AI) technology.
A recent press release highlights that many Canadian businesses are hesitant to adopt AI due to concerns about its safety, which could negatively affect their productivity.
“By tackling the societal, technical, and ethical challenges surrounding AI, CAISI will play a key role in fostering trust in AI technologies and supporting their responsible use across Canadian businesses and institutions,” the release states.
CIFAR was first identified by the Canadian government as a partner for the Pan-Canadian AI Strategy in 2017.
“The Canadian Artificial Intelligence Safety Institute will position Canada at the forefront of global efforts to manage AI responsibly, playing a vital role in building public trust in these technologies,” said François-Philippe Champagne, the Minister of Innovation and Science.
“In a fast-changing world full of uncertainties, Canadians can be confident that we will always take the necessary steps to ensure AI is safe, responsible, and reliable.”
With $27 million in federal funding, CIFAR will establish a research team to work with CAISI and the National AI Institutes (including Mila in Montreal, the Vector Institute in Toronto, and Amii in Edmonton) to collaborate with experts on addressing both short-term and long-term risks associated with AI.
“By partnering with research institutions nationwide, this initiative will expand on the work of Canada’s AI experts in areas such as detecting AI-generated content, evaluating advanced AI models, ensuring the safe use of AI in high-risk fields, and maintaining privacy in AI systems,” the release mentioned.
The launch of CAISI is a significant achievement, making Canada one of the few countries to establish an AI Safety Institute, alongside the United Kingdom and the United States.
“As the world faces the immense potential and challenges of AI, CIFAR is proud to play a leading role in Canada’s response,” said Stephen Toope, CIFAR’s President and CEO. “The investigator-led research initiative at CIFAR will leverage Canada’s strong AI scientific community to advance pioneering research on AI safety for the benefit of all.”
