ISED Publishes Voluntary Code Of Conduct For Advanced Generative AI Systems

On September 27, 2023, Innovation, Science and Economic Development Canada (ISED) published Canada's Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems (the Code). The Code is the result of extensive stakeholder consultations (as previously reported by the E-TIPS® Newsletter here), and is intended to provide voluntary guidance for those developing or managing the operations of generative artificial intelligence (AI) systems before future AI-specific regulation comes into force.

ISED acknowledged that while advanced generative AI systems have enormous potential for innovation in many fields, they also carry a distinctly large risk profile that is attributed to the broad scope of data used to train these systems, their wide range of potential uses, and the scale of their deployment. Specifically, they can present risks to health and safety, propagate bias, and have broader societal impacts.  To address and mitigate these risks, the Code identifies measures that organizations are encouraged to apply to their operations when they are developing and managing general-purpose generative AI systems.

Canadian firms who wish to commit to the Code should focus on achieving the following outcomes:

1. Accountability. This involves firms understanding their role regarding the systems they develop or manage, implementing appropriate risk management systems, and sharing information with others to avoid gaps.
2. Safety. Perform risk assessments on AI systems and implement mitigation steps to ensure safe operation before deployment.
3. Fairness and Equity. Firms should assess and test systems for biases throughout the lifecycle.
4. Transparency. Firms should publish sufficient information to allow consumers to make informed decisions and for experts to evaluate whether risks have been adequately addressed.
5. Human Oversight and Monitoring. AI systems should be monitored after deployment and incidents must be reported and acted on.
6. Validity and Robustness. Firms should conduct testing to ensure that systems operate effectively, are appropriately secured against cyber attacks, and their behaviour in response to the range of tasks or situations to which they are likely to be exposed is understood.

The Code already has a number of signatories, including TELUS and BlackBerry. In addition to their commitment to following the Code, the signatories promise to support the ongoing development of a robust, responsible, AI ecosystem in Canada by contributing to the development and application of standards, sharing information and best practices, and collaborating with others to support public awareness and education on AI.

Summary By: Victoria Di Felice