Privacy Supporting Innovation: Getting To Yes

On June 16, 2026, in collaboration with Sun Life and Environics Analytics, Dentons was pleased to host a roundtable forum, Privacy supporting innovation: Getting to yes. The event brought together Canadian thought leaders from government, regulators, industry and academia to examine the privacy and data policy challenges facing Canada and to identify practical, risk-based approaches that enable organizations to innovate responsibly with data.

We summarize the discussion and highlight some of the factors that emerged as potential “blockers” or “enablers” of “Getting to Yes.”

Setting the stage

“Institutional trust is a depleting asset.”

Ipsos Canada presented research on Canadian consumer attitudes toward data, privacy and institutional trust. Canada is experiencing a prolonged period of defensive consumer behavior, declining optimism and a relentless focus on price — described as a structural condition rather than a cycle. Net optimism regarding personal financial situations declined sharply over the past decade, with the 10-year outlook turning negative by 2025.

Ipsos distinguished between "old world" data use, focused on convenience and enhancement, and "emerging world" data use driven by economic and productivity gains — such as dynamic pricing, credit scoring, AI-driven hiring and personalized service access. This shift has significant implications for how Canadians experience and respond to data collection.

Central to the presentation was a distinction between two household types. "The Insulated" retain discretionary capacity and institutional trust, viewing technology as an opportunity. "The Exposed" are cash-constrained, with eroding trust, who exchange data out of economic necessity rather than genuine choice. Existing privacy frameworks assume the former, but reality increasingly reflects the latter — where consent is constrained and trust is a depleting asset. This dynamic risks producing a two-tier AI economy, widening the gap between those who benefit from innovation and those who bear its risks.

Summary of perspectives

“Privacy is the operating system of trust.”

1. Decision-making in a changing environment

While it is understood that we need data driven decision-making, effective policy requires choices. Needs are diverse across the country, and policy choices will involve competing priorities. The issues before us are complex and can only be addressed with discussion across government, regulators, industry, experts, academia and other communities. International collaboration must remain a focus.

2. Trust as a condition for privacy and innovation

Institutional trust should be treated as a depleted asset. Privacy consent was designed for a different environment, one built on assumptions of choice, reasonable discretionary capacity and reasonable institutional trust. That environment has changed, and there is now more constrained consent and decreased institutional trust. Consent frameworks should be stress tested and should not assume a high-trust population. AI can widen existing social and economic divides if they are not considered and built into system design. Poor communication is a major inhibitor of trust, and we must target communications efforts on building trust around innovation.

3. Responsible use of data

Data use must be reliable, fit-for-purpose and aligned with public trust. AI requires significant quantities of data, and unreasonable restriction could lower competitiveness. The key is to demonstrate public interest, responsible use and a clear purpose to innovate to secure social acceptability.

4. AI adoption and public confidence

AI job displacement remains a concern. There is a connection between trust, literacy and adoption, as when people do not understand AI or technology, they are less likely to trust it. Digital literacy efforts can increase trust, and adoption follows. AI may affect sectors differently as in some areas, AI is better understood as supporting existing processes rather than replacing workers, while in other areas, job loss and workforce impacts remain a concern.

5. Health care and consent

Health care is missing an automation layer. In other industries, value is created through automation, while in digital health care, efficiency can decrease if automation is not implemented properly. AI tools are most useful when they are trained on a specific task and designed for a specific role, as these tools can support professionals and help accelerate processes for patient care. Health care presents unique consent pressures, often seen as getting in the way, including when life-or-death decisions are being made, decisions involving children and other sensitive circumstances where consent may not feel fully voluntary, though a necessity to improve outcomes.

Hurdles to innovation in health care include data access, ethics approval mechanisms, security, safety, law, computation and governance. Privacy is the operating system of trust. As autonomy increases, existing frameworks may predate the technology operating inside them. Raising the bar on governance, transparency and oversight can help lower the barrier to responsible adoption.

6. Data protection

Data residency and data sovereignty are not the same. Data may be stored in one location while still being affected by legal or practical access risks elsewhere. Protecting data requires more than deciding where it is stored. Successful AI implementation depends on data, governance and trust. Trust requires transparency, fairness, accountability and responsible use. People want to feel protected, and while responsible innovation may slow growth somewhat, responsibility can support scalability and sustainability.

7. Compliance and accountability

Responsible innovation requires radical transparency, though transparency alone is not enough. There must be internal accountability and external accountability to regulators or other oversight bodies. There is a need to move beyond traditional compliance models: issues should be anticipated, relevant stakeholders should be consulted, and active frameworks should consider law, society, science and technical realities to ensure guidance is credible, practical and implementable.

8. Data for public and community benefit

Data can be used to make lives easier and create meaningful outcomes for individuals, communities and industry. Granular data can support local change by helping identify gaps, risks and opportunities for investment. It can also support community planning and local economic development when handled responsibly. The desired outcome and purpose should guide data use as data is there to be used, and it should be used for the common good.

However, accessing new data can also create conflicting narratives on the same issues. Privacy-enhancing technologies, deidentification, anonymization and controlled data environments can help respect privacy interests while still allowing useful analysis. Ongoing communication with end users and communities is important. As attendees shared some of their responsible data use cases, it highlighted the importance of open conversation about use cases and how we can effectively mitigate risk and deliver value and better outcomes.

Blockers and Enablers

Through the discussion, some factors were identified as potential “blockers” and “enablers” of “Getting to yes”:

Blockers

• Declining trust
• Lack of data integration
• Concern about AI job displacement
• Getting to “yes” requires effort – saying “no” is easy
• Cost of compliance
• Access to talent
• Cross-border restrictions
• Jurisdictional differences and legal uncertainty

Enablers

• AI Literacy
• Social acceptability through demonstrated necessity
• Privacy-enhancing technologies (anonymization, pseudonymization…)
• Data governance and accountability
• Democratizing data access
• Government leading by example
• Support from corporate leadership
• Interoperability
• Government supporting investment
• Data-driven decision making

In the end, though, privacy done right is increasingly seen as an enabler to innovation allowing us to go faster, suggesting we don’t have an innovation problem in Canada, but rather a trust and execution problem. The work of getting to yes starts now – in how we design, govern and actually deploy.