Policy to practice: simulations can bridge the government AI skills gap
While the Australian Government’s 2024 ‘Policy for the responsible use of AI in government’ mandated accountability positions and statements be in place by February 2025, many anecdotes suggest the policy landscape remains in partial formation rather than mature operation. This leaves many arms of government confused or complacent about how to safely understand and harness the new wave of AI tools.
This state of affairs was evidenced at Canberra’s recent Public Sector & Government Communications Week conference in May, where event Chair and Deakin University academic Ross Monaghan heard some very familiar concerns around the conjoined challenges of artificial intelligence and stakeholder engagement within government agencies.
Delegates at the conference vocally fretted over their departments’ reticence to sanction (far less optimise) the use of AI technologies, vented about feeling ‘left in the dark’ and expressed worries on how to tackle real-life incidents or crises where AI assets are central to the threats presenting. Interestingly, similar concerns were also voiced over the use of social media channels, which have been around for over 15 years.
“Public sector staff feel the whole AI issue is getting away from them and they want tools, training and support to ensure they’re not left behind in the race to use AI safely,” Monaghan said.
To address this real disparity, Monaghan has integrated his academic research on AI into a series of workshop-based events, using a locally built simulation software. Monaghan insists that interactive software (powered by AI assets and content) can better help government departments, directors and employees to understand the pros and cons of AI by realistically experiencing them.
Let’s be honest: the problems with AI can be manifold — from policy gaps and service delivery implications to intellectual property issues and misleading information generated by AI bots. Still, the majority of us recognise that — far from going away — AI technologies are already in common use in many private sector settings and delivering benefits including data research, improved efficiency, innovation and work output quality.
Monaghan summarised his public sector peers’ concerns as falling into three principal categories: policy gaps; skills development; and applying knowledge.
Policy gaps
Research from mid-2024 by the Australian National Audit Office (ANAO) showed that of the 56 entities using AI that year, just 36 had AI governance policies and only 15 had AI assurance policies. Another paper by the Community and Public Sector Union (CPSU) indicated around 60% of organisations were witnessing unauthorised AI use while only 35% of those entities officially permitted it. Other feedback suggests a general lack of guidelines or formalised training on appropriately deploying AI tools.
Skills development
While the private sector seems to be embracing AI ‘smarts’, many government employees feel they’re missing out on crucial learning skills to help them understand and master AI, while departmental and political leaders debate and deliberate on the topic. If departments aren’t investing in staff training for AI, they’re actually creating a two-tiered workforce where public servants can become disengaged due to their perceived skills disadvantage.
Applying knowledge
How can workers mitigate risks from tools they’re banned from using? Risks or threats emerging from AI actors — algorithmic bias, cyber risks, deep fakes and synthetic (false) data — can be impossible to handle when the tools and tech that create them are largely alien to the workers tasked with containing them. But it’s not enough to intellectually understand what these tools can do; you need to actually practice how to use them; which is where interactive learning tools and simulations come to the fore.
Like it or not, use it or don’t, artificial intelligence is impacting the lives of citizens and stakeholders right across Australia.
Given this exponential rise, Monaghan feels it’s time for communicators, policymakers, lawyers and IT specialists in local, state and federal government to get around the table — not just to chat, but to interactively prepare and train for how to master AI. Monaghan — a communications teacher and trainer of 20 years’ standing — insists that simulation learning isn’t some kind of academic game but is a realistic and needed training ground for coming to grips with AI, at all levels of government.
Interview: Phil Zammit, Avaya
For our summer Leaders in Technology series we are discussing AI, system complexity, skills and...
From training to collaboration: how XR can transform public sector operations
Extended reality (XR) has emerged in recent years as a powerful tool for public service...
How financial automation bridges the public sector resource gap
The growing resources gap in finance and technology is a key challenge in the public sector.