AI-powered Government: How the Natural Resources Canada Digital Accelerator brings data science into public services
Governments are keen to deploy the latest advances in data science and artificial intelligence to optimise and fine-tune many of the workings of the state, from citizen-facing services, to planning and policy.
Nevertheless, they face daunting challenges in doing so successfully, from battling for AI talent with the private sector to managing the risks that come with new technologies. Two years since its formation, the Natural Resources Canada (NRCan) Digital Accelerator (DA) shows how institutional innovations can help the public sector to responsibly use frontier technology to benefit governments and their citizens.
NRCan, the department responsible for managing Canada’s abundant natural resources and charting the nation’s sustainability transition, was the launchpad for one such accelerator initiative. The Digital Accelerator’s goals are to seek ways of using data science and AI to lower costs, accelerate productivity, and improve the accuracy and precision of the department’s work – which includes forecasts, predictions and mapping of resources.
The accelerator has four key aims in its strategy, as articulated by Dr. Vik Pant, former NRCan Chief Scientist and Departmental Science Advisor. First, build capacity by gathering a pool of highly skilled data scientists with backgrounds in areas including mathematics, statistics, and computer science, and with direct expertise in AI. Rather than being a standalone ‘turnkey’ unit to deliver projects, a team of highly complementary, synergistic skills has been assembled to provide cross-functional support to science and policy experts across NRCan.
Capacity was a necessary but not sufficient step to success. The second goal is to identify specific ways for AI to deliver additional value. NRCan was already using many technical and scientific tools in its work in forecasting, simulation and prediction, all of which are domains that today’s AI and data science techniques are useful for.
However, the key is to complement existing methodologies rather than replace them. Much of the similar work has been happening through conventional methods. NRCan understood the potential that the latest AI, data science, and machine learning bring to complement existing models. The current roster of projects includes applying machine learning to classify and analyse satellite imagery, which helps in areas like flood and fire mapping; detecting mineral deposits using deep learning; deploying AI to improve access to groundwater; and using big data to optimise electric vehicle infrastructure.
The third goal is to build momentum on a department-wide scale. The DA team wanted to make sure there was a cultural aspect of the work, to elevate the collective digital quotient of the organisation.
This cultural transformation is achieved by everything from “lunch and learn” and “ask the expert” sessions to networking and knowledge-sharing exchanges. Most recently, the Digital Accelerator is lending support to the highly anticipated first Government of Canada Public Service Data Challenge, led by NRCan in partnership with Statistics Canada and Global Government Forum.
The Data Challenge will bring together diverse talent from across the public service, regardless of their background or expertise, to showcase innovation and mobilise data to improve citizen and employee services and support evidence-based policymaking, evaluation and analysis. The Accelerator will provide consulting and mentoring support to participating teams. Call for proposals will open in April 2022 and the program will run until Feb 2023.
The fourth goal is to pursue strategic partnerships – including alliances with software firms, universities and research institutions – to make use of the frontier knowledge that is outside of government, which is moving fast. NRCan is working closely with the likes of Microsoft and Google, which are world leaders in AI.
Bringing innovation across government
The NRCan Digital Accelerator has made waves across the wider Canadian government.
The challenge comes with coordination of efforts to avoid duplication or even working at crossed purposes. Governments are not only battling for AI talent with the private sector but among themselves as different departments seek to offer the best terms and perks for digital and AI talent.
The beauty of accelerators, and the cluster model, is that government can instead form a talent pool to focus on cross-cutting questions that different departments grapple with. Algorithms can bring huge improvements in the speed, cost and precision of government service provision, but they can misfire, and the public rightly demands to know that decisions, which impact them, are based on fair, transparent and explainable reasons.
Paul N. Wagner, Assistant Deputy Minister of Strategy and Transformation at Treasury Board Secretariat, describes the work done by NRCan’s Digital Accelerator team as a “great framework to build from. Now, we want to get enough momentum behind the idea, and enough leaders within government to lead these accelerators, to start to benefit more from this experiment”.
With much success achieved in a short period of time, the Digital Accelerator team looks forward to continuing to build on the energy and foundation that has been built over the last two years.
About Partner Content
This content is brought to you by a Global Government Forum, Knowledge Partner.
Related Posts
