Liberate our Canadian data for AI adoption

As an organization that drives the democratization of artificial intelligence in the Canadian private and public healthcare sector we understand the Canadian AI paradox. We are and aspire to remain a global leader in AI developments and breakthroughs, however adoption of AI in our businesses and in the public domain evolves slower than expected.
The Canadian eco-system for fundamental AI research, the development of AI talent and the support for startups in the domain is well developed, but Canadian companies are embedding AI in their operations and strategies at lower pace to allow to have Canada remain its leadership position.
The series of reports released by Deloitte over the past year provides unparalleled insights in the requirements for the eco-system to further evolve and reach global leadership status.
“As a country, we have the research strength, talent pool, and startups to become a leading AI supplier, but that’s not enough if we want to lead in an AI-driven world. Our ambition as a nation should be to shape what that world will look like. True leadership is required―that means taking steps now to establish a world- class AI ecosystem in Canada.”
One of the reports: Canada’s AI imperative Public policy’s critical moment focuses on some fundamental changes that are required to push the AI movement forward. Unlocking the value of data is a crucial part of that. Below a fragment of the report that can be consulted at
“If data is the “new oil,” as The Economist and business leaders suggest, then giving Canadian businesses more data and regulatory certainty concerning their data is crucial to unleashing the demand for AI. (6 )
Why this matters for AI prosperity
Good data is what makes good AI possible. If AI is going to drive our economy, Canada needs to increase the quality and quantity of public and private data available to researchers and businesses. This is an urgent issue for Canadian competitiveness, as the quality of algorithms is directly tied to the quality and quantity of the data with which the algorithm operates. Companies without usable data can’t embed AI deep into their operations and strategies; they’d be limited to more superficial uses.
But prosperity from AI depends on more than the quantity of data or getting it into the right form. Our past research shows that Canada needs leadership from business and government to build trust in how data is collected and used.(7) We need to unlock the value of data the right way to make it available for AI development while preserving the privacy and trust of citizens.
Public policy recommendations
Reform intellectual property law to respond to machine learning: Intellectual property law needs to provide clarity for practices such as data scraping (extracting data from websites and other sources into a machine-readable form). Data scraping is key for AI because it facilitates the creation of large amounts of datasets for algorithms quickly and easily, but Canadian law does not currently address data scraping, putting it in a grey zone and slowing adoption. Deloitte’s environmental scan of national AI strategies found that other jurisdictions are moving to create and adopt explicit text- and data-mining exceptions to copyright law―a move the Canadian government should emulate. (8)
Make public data available for commercial use: Federal, provincial/territorial, and municipal governments in Canada can help spur innovation by making public data available in machine-readable format for commercial purposes. Making publicly held data such as utility, transportation,and health-care data available is a feature of national AI strategies in France, Germany, and China, and a focus for the European Commission.(9)
Separate from privacy laws, the federal government and the Government of Ontario are both developing data strategies to maximize the value from data while, at the same time, limiting the inherent risks that come with data sharing (data sharing being necessary to maximize value). These strategies should set clear goals and boundaries for data governance and chart the underlying principles that will guide how Canadian companies are allowed to collect and use data. Government data policies should respond to low public trust in data handling and use, and recommend best practices to allow consumers to be genuinely in control of their data.
Lay the groundwork for data trusts: Data trusts— fiduciary trusts that hold and can make decisions about data on behalf of the individuals who generate the data—are a promising practice, especially for the management of public data. But awareness of data trusts is low; for these trusts to work, people need to understand their role in managing and protecting their long-term data rights.(11) Governments should also issue guidance and clarify the legal environment regarding data trusts―for instance, diligence standards for trustees―to enable and standardize their use.(12)”
MUUTAA helps their partners in the healthcare domain understand where significant advantages can be unlocked, given the availability and accessibility to data. Early AI success with less complex models, will help the organization understand the importance of data and see the value of applied AI. A first step to a profound AI strategy.
*Picture: Macleans

Leave a Reply