It lost students and $36 million out of an already relatively small budget last year, but the University of New England appears to be on the cusp of transitioning from ugly duckling to cygnet thanks to two little letters (yes it’s ai).
Chris Moran took over a beautiful but long-troubled institution when he moved into the corner office at Booloominbah – the historic country house that serves as the University’s Armidale Chancellery.
Staring down deficits and determined to reverse the decline in demand, the new VC initiated a range of new approaches, but none as quietly transformative as sticking all ai innovation under the creative gaze of Associate Professor Aaron Driver, who brings not only a world of varied experiences demonstrating a capability to engage people, but also a devotion to out-there names.
Lab Next 70 was the name given to UNE’s Applied Innovation Lab. Within less than 12 months, A/Prof Driver and his team have achieved results that are quite astonishing – closing in on 100% of staff using AI tools in a system built around staff empowerment called Madgwick, after the founder of the University.
Speaking publicly for the first time about the University’s implementation plan at the Future Campus HE FEST conference, A/Prof Driver is quick to note that they may never get every single staff member on board, but outlined the approach that has turbocharged adoption.
“Many universities are stuck in AI pilot purgatory,” A/Prof Driver said.
“We wanted sovereign control of data but also a choice of models, tools and tasks – to embed intelligence into the organisation.”
Rather than grapple with the costs of individual licences for generative AI models for each staff member, the university partnered with Newcastle-based developers Simtheory to develop a whole of enterprise solution with local hosting and support. By paying for usage rather than seats in an LLM, the university is able to afford widespread adoption – at a cost of less than $250,000 for the first three quarters of the calendar year.
UNE have around 1,200 staff but already have 1,197 users on the Madgwick platform – which has been set up so that A/Prof Driver’s team can learn from patterns of usage and demand. Staff have so far created almost 3000 assistants, with 126,000 interactions with AI agents over the 30 days to mid September. Just 14% of engagement involves ChatGPT, and the AI team provide regular advice on the best tools to use for a variety of use cases.
This is an extraordinarily high rate of adoption over the course of less than 12 months, and is likely to turn UNE from a university that has been easily ignored, into an institution which could become an international benchmark for change.
It’s still early days though. The team are developing a series of lessons to guide staff through the basics of choosing AI systems and creating AI assistants.
“Our aim is to develop a fully thriving AI literate workforce,” A/Prof Driver said.
“What I am seeing anecdotally is everything is being done more quickly and efficiently.”
“We don’t have a replacement mindset, we have a deployment mindset, if you can free up some time by using AI, then you can deploy that time to do more.”
The team are looking at opportunities to leapfrog the university’s technological debt – the challenges of grappling with ageing legacy systems for enrolment and admissions may be bridged with a layer of AI assistants, A/Prof Driver said, but the jury is still out as to how far the issues of systems that don’t talk to each other can be resolved.
“We are hoping we can leapfrog a lot of the tech debt through AI, we are not sure, but we are hoping we can,” A/Prof Driver said.
In the meantime, UNE staff are using more than 100 AI tools to revise policies, update themselves on procurement and travel policies, or draft key documents. The sector in Australia and internationally are bound to be closely watching progress.
“I go to events and people say I love Madgwick, I love talking to my assistants,” says A/Prof Drive
“Who says that? Especially about tech?”
La Trobe University Vice-Chancellor Professor Theo Farrell told the HE FEST audience that one of the keys to AI adoption across the institution was to ensure staff felt they had permission.
“Using AI is not cheating, it’s clever,” he said.
Professor Farrell outlined how La Trobe had initially introduced a three-year blueprint for AI adoption, starting with productivity, then moving to differentiation and disruption.
“We realised at the start of our trial we don’t have to bother about productivity – 81% of people were already using AI, we didn’t have to tell them to use it,” he said.
As a result, the university dropped the productivity plan and moved on to the next phases of implementation – underscoring the critical importance of being flexible in implementation.
“We are experiencing something that looks like the industrial revolution in terms of how it will change our existence. The simple fact is we don’t know yet where it is going to land,” he said.
“Anyone who tells you exactly where the world is going to be in a decade is selling you snake oil.
“Whether you like AI or not is absolutely immaterial, because the level of investment is so vast, there is so much business muscle and investment behind this, this thing is like a locomotive coming down the tracks.”