Communique
The NSW Government held the AI Thought Leaders Summit on 29 November 2019. The Summit was opened by the Minister for Customer Service, the Hon Victor Dominello MP, and the Secretary of the Department of Customer Service, Emma Hogan.
The Summit was chaired by the NSW Government Chief Data Scientist and Chief Executive of the NSW Data Analytics Centre, Dr Ian Oppermann, and hosted by the NSW Government Chief Information and Digital Officer, Greg Wells.
Invitees included academic, government and legal experts. They heard and commented on detailed case studies and presentations that focused on AI project successes, risks and mitigations, ethical considerations, data governance, and building digital capability in the NSW public sector.
The Summit provided an opportunity for attendees to make recommendations to ensure that the NSW Government’s AI Strategy contains clear commitments that will advance its maturity in this emerging technology. It was noted that the Strategy will be focused on service delivery and better decision-making, and that attendees will be further consulted after the next iteration of the Strategy is finalised in the coming weeks.
Attendees received pre-reading discussion papers based on earlier consultation with industry, academic and government representatives about AI, its risk and benefits. Key themes that emerged were the need to build public trust in AI and how this might be achieved; building NSW Government capability in data use and sharing; ensuring our public servants are well-equipped to make informed decisions about emerging technologies; developing a culture of innovation and collaboration with universities and industry; and reviewing the procurement framework to facilitate uptake of AI and other digital technologies.
As part of the pre-reading, draft NSW Government principles for the use of AI were distributed to attendees:
-
Community benefit - AI should deliver the best outcome for the citizen, and key insights into decision-making
AI must be the most appropriate solution for a service delivery or policy problem. It should always be considered against other analysis and policy tools. AI should be the solution that maximises the potential benefit for the customer and for government. It should not be used where there is not a clear case for doing so.
AI will be used when it provides the best option for service delivery or informing government decision-making.
-
Fairness - Use of AI will include safeguards to manage data bias or data quality risks
The best use of AI will depend on high quality and relevant data. AI solutions that rely on poor quality data may result in sub-optimal project outcomes and recommendations. Algorithms that contain systemic and repeatable errors may lead, even being based on high quality data, to prejudiced decisions or outcomes.
Safeguards will be put in place to ensure optimal data quality and objective analysis.
-
Privacy and security - AI will include the highest levels of assurance
NSW citizens must have confidence that data used for AI projects is used safely and securely, and in a way that is consistent with privacy and data sharing requirements. Any project outcome will be undermined by lack of public trust if there is any risk of a data breach or that personal data could be compromised.Privacy and personal information assurance mechanisms will be at the core of all AI projects.
-
Transparency - review mechanisms will ensure citizens can question and challenge AI-based outcomes
Not only must the people of NSW have high levels of assurance that data is being used safely and in accordance with relevant legislation, they must also have access to an efficient and transparent review mechanism if there are questions about the use of data or AI-informed outcomes. The community should be engaged on the objectives of AI projects.
AI-informed projects, decisions and outcomes will be explained quickly and transparently.
-
Accountability – decision-making remains the responsibility of organisations and individuals
AI is a powerful tool for analysing and looking for patterns in large quantities of data, undertaking high-volume routine process work, or making recommendations based on complex information. However, AI-based functions and decisions must always be subject to human review and intervention.
Service delivery and policy design which leverage AI will be regularly reviewed. AI will be used as a tool to assist policy making and service delivery, rather than be the ‘policy maker.’
While there was high level agreement with the principles, and the overall policy approach, the Summit emphasised that a principles-based approach would not be possible without careful application, and without an assurance and review mechanism.
In relation to the development of an AI User Guide for NSW Government agencies, the Summit noted that the application of principles should be clearly articulated throughout the guidance material, so that those preparing a case for the use of an AI solution can more easily make an informed decision about whether that solution will meet NSW Government standards.
While it was accepted that AI can be a powerful tool for problem-solving and informing decision-making, the Summit emphasised that accountability must remain with agencies and government. It is important that any technology that impacts our communities be subject to careful and regular human review. That is, AI will provide support for government decision-making, and government must retain accountability for decisions about service delivery.
The Summit noted the work that the Public Service Commission is undertaking to build digital capability across the sector – for senior -decision-makers, project leaders and those specialising in data analysis. The Summit advised that, in addition to digital skills, public servants need to continue to develop awareness of where AI is the right solution (or is not), how to ensure the right questions are being asked and that the outcome is ethical and based on quality data. In summary, the Summit emphasised that government needs to make quality, ethical decision, irrespective of any technology considerations.
A range of issues were raised in relation to the question of building public trust in AI and digital technologies. The Summit emphasised the need for privacy considerations to be paramount, and that there be speedy redress mechanisms in place in the event that the community wants a review of a decision. Further there should be a commitment that no-one should be any worse off after a technology-informed decision.
The Summit closed at 4.30pm. Attendees noted that they would be sent the next iteration of the draft Strategy for their consideration and further feedback and that the NSW Government would release the final document by Q1 2020.