Enhancing Service Delivery in the Public Sector: Kate Burnett-Isaacs

Kate Burnett-Isaacs, Director of Data Science at Infrastructure Canada (INFC), is steering the department towards a future where artificial intelligence enhances public service delivery. In her upcoming talk at CDAO Canada Public Sector, she will detail how the ethical and practical applications of AI are transforming her department’s operations

Infrastructure Canada’s mandate is to improve the quality of life of Canadians by creating world-class, modern public infrastructure. As a government department, it invests in enhanced public infrastructure to build inclusive, sustainable, and prosperous communities that meet the needs of all Canadians.

The department’s remit includes reducing and preventing homelessness, transit, resilient and sustainable infrastructure, and community building. With its recent emphasis on integrating artificial intelligence (AI), INFC is striving to modernize how public services operate, making them more efficient and responsive.
As the Director of Data Science, Kate Burnett-Isaacs and her team are leading INFC’s charge to harness AI in order to tackle some of Canada’s significant challenges, including ensuring affordable housing, as well as effective infrastructure planning and environmental sustainability.

Kate’s presentation at CDAO Canada will explore how Infrastructure Canada is implementing an AI Framework to pave the path for ethical and responsible use of AI in practice. With guidelines, tools and project governance designed to address business challenges and opportunities first, learn how AI will be used in a practical, efficient, and sustainable manner to improve the efficiency and accuracy of INFC’s operations. 

Check out the full agenda and register for the event here.

How does the practice of data analytics in the public sector differ to the private sector?

Kate Burnett-Isaacs: Any work done by the public sector is to improve the lives of Canadians and those living in Canada. For us that means providing services and solutions through Infrastructure Canada funding programs and initiatives. Our mandate now includes housing, aiming to enhance Canadian lives through public investments in housing, homelessness, infrastructure, and communities. 
My role involves supporting reproducible analytics, advanced analytics, and AI to enhance decision-making, improve policy programs, funding, reporting outcomes, and increase transparency in our operations. Ultimately, we want to serve Canadians directly using data to do in the best and most efficient way possible.
We are focused on organizing, managing, and utilizing data to build insights. As we've expanded to include green infrastructure, climate-resilient infrastructure, homelessness, and housing, it's even more crucial to make data and analysis easier and more accessible.

Moving to the topic of your upcoming talk at CDAO Canada Public Sector, how do you approach the challenge of implementing an AI framework while laying the path for ethical and responsible use of AI in practice?

Kate Burnett-Isaacs: My main goal is to treat AI as a tool to solve problems, not to throw AI at every challenge. We assess the business needs first—identifying pain points and opportunities—then determine if AI is the best solution. We're focused on bringing AI into practice, making it tangible and practical for everyday use while ensuring people are accountable for AI outputs.

There are several areas, especially in decision-making processes, where AI can enhance efficiency and accuracy. For example, using AI to forecast grants and contributions expenditures has increased accuracy and aided our finance team in their forecasting tasks.

Another area is in synthesizing information from parliamentary committee minutes faster and more efficiently, making essential data more accessible. 

How do you address the human aspect of integrating new technology like AI?

Kate Burnett-Isaacs: We've worked hard to position our AI framework as a practical tool, akin to a search engine, making it accessible and understandable. Emphasizing data quality is crucial—ensuring data is available and of high-quality builds trust in our data and the algorithms we use.  INFC staff are to treat AI as an important tool. AI outputs must be reviewed by staff who are ultimately responsible for filtering the information and making decisions.

How are you building capabilities for implementing practical, efficient, and sustainable AI practices?

Kate Burnett-Isaacs: We emphasize a 'data pipeline' from collection to dissemination, involving end users in every step. Our data stewards' community of practice ensures data quality and ownership. 
By removing the 'black box' aspect of AI and involving clients throughout the development process, we ensure the outputs are useful, understood, and trusted. So, throughout the entire process, there's checks, there's validation and then training in a wide variety of forms.

What's your approach to measuring success and proving the value of AI projects?

Kate Burnett-Isaacs: Measuring time saved and accuracy of models compared to human effort and error is crucial. We also consider qualitative aspects, such as how frequently AI tools are used and whether they're integrated into regular processes. Our aim is to transition from manual, cumbersome processes to more efficient, AI-driven operations.

Considering the ethics of AI, have you had to adjust your approach for certain projects?

Kate Burnett-Isaacs: We always consider ethics from the start, guided by the Treasury Board's directives and our own framework. For instance, we are exploring how to use Large Language Models to pull in and synthesize data, ensuring ethical use and thorough evaluation. However, we keep a very close eye on all the data going in as well as the output. 

Lastly, how do you see the data and analytics function impacting the public in Canada over the next few years?

Kate Burnett-Isaacs: Our immediate focus is on internal operations, using data to inform decisions on policies and programs. Eventually, we aim to use our insights to directly reach Canadians, improving access and processes to enhance our connection with the public. The long-term goal is to use data and technology to better serve and engage with Canadians.

Want to learn more?

Kate Burnett-Isaacs is speaking at CDAO Canada Public Sector, kicking off on June 18^th, 2024. Join her and many more world-class data and analytics leaders at CDAO Canada Public Sector to learn about the trends driving value creation in public sector data and analytics. Register to attend here.