MicrotechDPS Chief Operating Officer, Joel Steers, recently joined a panel to discuss this critical topic at the highly anticipated NESA event. We sat down with him at the summit to discuss all things AI in an Australian employment services context.
“This is an exciting time for employment services. We rarely have a chance to upskill our entire workforce at one time, and AI is presenting that moment as we speak. The workplace productivity opportunity right now is immense. It’s a meaningful moment,” he says.
The challenge for employment services providers
But for employment services, Joel notes that there is an added challenge when it comes to using AI.
“AI operates within a heightened regulatory and ethical environment. Providers work at the intersection of government policy and deeply human outcomes, where decisions, recommendations and system‑driven bias could materially affect a person’s employment pathway.
“In Australia, this means AI use is guided not only by general technology best practice, but also by whole‑of‑government AI principles and DEWR‑specific expectations that extend to third‑party providers.”
Joel highlights that these guardrails do not exist to necessarily slow innovation. Instead, they shape where AI adds real value for providers now.
“The real opportunity for employment services in AI right now is about reducing low‑value administrative effort, supporting staff capability, improving consistency, and protecting the human relationships at the centre of the work.”
“The challenge for employment services providers is in realising the value in a way that is responsible, sustainable and impactful. And that really starts with solid foundations,” Joel explains.
What to consider when implementing AI in employment services
He highlights four key areas for employment services to consider:
- AI strategy: Do we have an AI plan or strategy that is realistic and works towards our business goals?
- Known and appropriate tools: Selected based on business needs not just the latest tech at that time
- Development capability: Do we have the people in the business, or partners to assist us, that can develop and maintain these tools?
- Technical foundations: Do we have the foundations of our AI correct? Can we be confident in protecting our sensitive data from inadvertent AI use?
AI strategy for the employment services industry
Joel says that when creating your AI strategy, it’s a good idea to start with realistic, measurable and smaller goals that align with your business objectives.
“Start small, in areas where you will see outcomes quickly, such as personal productivity improvements for staff and areas that take away the lowest value work. This can help to get people on board early and make noticeable inroads quickly.
“A clear strategy is, of course, an important first step for your business to think about how best to use AI, but it will also help you take your team on the journey with you. An engaged team is critical to its success.”
He says it can help to look for early adopters amongst the team; people on the floor doing the job who are on board with the AI transition. He suggests bringing them in early and including their voices in the planning process.
“These people can become AI champions amongst your team, helping others realise the value and get on board. If you don’t get people on board, they won’t use the tools and even the best strategy will struggle.”
Joel adds that a defined strategy and plan will also help to avoid ‘AI sprawl’.
“It’s not uncommon to have deployed individual tools over time and end up with what we call ‘AI sprawl,” says Joel.
“There’s a tendency to make AI decisions based on the latest shiny, exciting tool. But what this means is you essentially have a series of potentially expensive tools that don’t work together or integrate with your workflows and processes, so you aren’t able to realise the full workplace productivity gains.
“It can also create more risk and governance issues as you don’t have clear guardrails for each tool or how it intersects with your data.”
He says that the key to avoiding or remedying ‘AI sprawl’ lies in understanding the organisation’s goals for the use of AI, adopting an AI plan and then choosing fit-for-purpose tools.
Choosing the right enterprise AI tools as an employment services provider
Joel says that rather than focusing on their tools themselves, start with a clear understanding of where you can make the biggest efficiency gains, then work out which tools are best suited.
He highlights that some of the biggest gains he sees among employment services clients are around client onboarding, maintenance of documentary evidence and end-to-end compliance.
“We’ve seen highly successful digitization of these elements into a good document management platform which makes the capturing of the information easier, as well as easy retrieval and compliance reporting possible.
“Business process automation has been around a long time and has made great gains. AI is now the next level of that taking things beyond rule-based, repeatable workflows to interpret, extract, classify and search by context rather than keywords, or even summarise or answer questions based on stored information.”
Agentic AI is a key part of the solution for many, says Joel.
“We’re seeing a rise in adoption of natural language agents to answer staff questions. A question that might have involved seeking out advice from their manager or sifting through lengthy documents can now be asked and answered in seconds using your stored datasets, like guidelines and policies. This means staff can ask questions in natural language and receive fast, accurate and detailed answers to help inform them on what to do or how to act.”
Guardrails for AI implementation in the employment services industry
Joel highlights that it can be easy to get excited by AI’s potential in the early stages, but it’s important to go into any AI project with plans in place to protect your data, your team, your customers and your business.
“It’s not the most exciting part of your AI project, but it might be the most important. Understanding how you are going to use the technology, what walls you will put up around your data and what expectations you have of employees when it comes to the technology is the first step to ensuring you derive meaningful value. If you skip these steps, fitting them in retroactively can be challenging as AI tools can take on a life of their own if you lack appropriate guardrails from the outset.”
He notes that the Voluntary AI Safety Standard, now reflected in the Australian Government’s Guidance for AI Adoption, gives providers a practical foundation for governing AI without unnecessarily stifling innovation.
“Used well, it acts as a decision‑making framework, helping organisations to be clear on purpose, accountability and risk, while prioritising AI use cases that support staff, protect trust and strengthen service delivery outcomes.”
He adds that as the government looks to set regulation in a fast-moving space, now is the perfect time to set your organisation up for success.
“While we wait for the government to approve requests and provide more clarity to employment services providers, you can start working on your own strategy and guardrails so you’re on the front foot when the time comes.”
Effective AI deployment in the employment services industry
Once you have set your foundations and guardrails, it’s important to think not just about what you will deploy, but how you will deploy it.
“Organisations don’t need to begin with the most complex area – they can start small on simple personal productivity gain, taking away the low-value work so your staff can start working on more meaningful tasks right away, and work one on one with the early adopters to get things moving.”
Joel says organisations often succeed at this initial deployment, perhaps with support from the vendor or a third-party support team, but things can start to fall down when it comes to ongoing development and maintenance.
“The excitement and momentum of a new AI tool can quickly dissolve if you don’t have support for ongoing development and maintenance. These tools aren’t set and forget, you need to think about how they will evolve with your business – and who will help you ensure they stay relevant to your business.”
He also points to ongoing training as a key area of investment.
“Training is imperative. Like the tools themselves, training your staff in AI is not ‘one and done’. It is about ongoing capability uplift that will see both business and team members thrive over the longer term. It’s about helping them to not just understand what the tools can do but how to leverage them to achieve significant workplace productivity gains now and into the future.”
He highlights that successful deployment can be about having the right people in your business to maintain and develop the tool and train your teams. Or it can be about outsourcing to the right partner.
“It’s not feasible for every organisation to employ full-time development, training or maintenance staff. And so this is something we are doing more and more for clients, providing that support to maintain and further develop an AI platform so that productivity gains are sustainable and long-term return on investment achieved.”
Looking at your AI options as an employment services provider? Talk to MicrotechDPS first. We can help you put the foundations in place, deploy successfully and ensure long-term return on investment.
