photograph: RNZ / Unsplash
Twenty-three years ago, Karker's small, raised population was transported to Zealandia Sanctuary, shoved into the hill above central Wellington.
Their reintroduction has been an astounding success, and today Karker's ventures are far beyond the predator barrier.
“The population has exploded,” says Dr. Andrew Lensen.
More Karker, more problems.
“Karker likes to start nests in the attic, pull people's ditches apart, tear their favorite roses, like lead and accidental poisoning from bait stations,” says Andrew.
There are so many karkers now that combined with the ability to roam widely in the distance, it is becoming too difficult to track birds with banding and bullet age.
But ecologists want to understand how they are riding, where they are hanging out, and how human conflicts are occurring.
Input: ai.
Follow the changing world Apple, Spotify, iHeartradio, Or where to listen to the podcasts.
Andrew is a senior lecturer in AI at Telenga Waka Victoria University in Wellington. One of several applied AI projects he has worked on involves training an AI model to communicate one of Wellington's feathered residents, separate from others, using a distinctive Kākā feature.
“The pattern of the beak, the curvature of the beak, perhaps their posture, the wounds on the beak, things like that.”
He and his collaborators piloted the idea using a feeder box like a photo booth with gopro inside to take headshots of the birds as they feed. Now they're expanding, Andrew says. “We're getting more cameras. We're using trees and things to move beyond the feeder box on a more realistic background.”
Karker is currently a common sight in Wellington City.
photograph: Judy Lapsley Miller
This is just one of many amazing ways in which AI is used in New Zealand.
Andrew believes this lack of trust is partly attributable to the lack of understanding of AI, how it works, and its various forms and applications.
His own research focuses on explainable AI aimed at making machine learning systems more transparent.
Dr. Andrew Rensen, Senior Lecturer of AI and Program Director of AI at Victoria University Wellington University
photograph: RNZ / Claire Concannon
Artificial neural networks are based on the infrastructure of our own brain (the original neural network) and how we can make decisions based on the way it flows through it. The data is placed in this artificial network and processed through various connected nodes before getting answers on the other side.
Deep learning neural networks are a complex version of this artificial intelligence, with many layers of connected nodes. They could be very good at asking them to do and spit out great answers, but the way they actually do this is a “black box,” says Andrew.
“It's very difficult even for experts who learn in depth to look at you or me or those numbers and understand what it's doing, for example, if they're given a picture of an animal and say it's a bird or a cat.”
Explanatory AI can use complex models that you can understand, or explore more complex models, and get clues about how they work. But these simpler models may not work well, Andrew says. “There's this trade-off, but there's actually a high-performance model and a model we can trust.”
When AI is used in applied research, that trust is extremely important, he says. “The worst thing you can do as a computer scientist is to say, 'Hey, I made you this cool, just trust it, it's fine.' ”
Verian's 2024 “Internet Insights” report showed that New Zealanders are more interested in AI than they are excited about it. In this survey of 1,000 people, 68% were very or very concerned about the use of AI for malicious purposes, while 62% were concerned about the inadequate regulation and law. Less than half (49%) were very or very concerned about AI's impact on society, with only a quarter of respondents saying they knew a significant amount of money about AI.
However, AI is becoming inevitable soon. It is reshaping our online landscape, and its use is also increasing in the business sector, including New Zealand's public services and RNZ.
Can it also solve some of our most difficult questions?
AI and farming
Professor Bing Xue from Wellington University, Victoria, is part of a large team aiming to use data science and AI to increase aquaculture productivity in New Zealand.
The $13 million seven-year research project (given in 2019) aims to optimize the agriculture of green wet mussels, king salmon and oysters, and use a variety of tools and means to do so.
This includes informing farmers whether the Massar buoys have been lost due to the development of detection systems and alerting systems. This eliminates labor-intensive manual checks. They are also collecting data on the mussel lifecycle, from spat sourcing from 90 miles of beaches to fully grown adults, to investigate how this growth can be made more efficient.
This project combines AI with real-world knowledge. Bing, it's important to go out on the farm and talk to marine biologists, engineers, engineers and farmers. “I learned something I don't know when I work [with] “Lab AI,” she says. “If you want to solve a real-world problem, you need to work with experts in that particular field.”
Dr. Xue Bing at Mussar Farm research site.
photograph: supply
Measles modeling
Her goal is more about preparation than productivity, but so is Dr. Fiona Callahan.
Fiona is the chief advisor of epidemiology for public health agencies within the Ministry of Health. She works with PHF Science infectious disease experts[ (formally ESR) to model an outbreak of measles in different regions in New Zealand.
Our last large measles outbreak was in 2019, when more than 2000 people were infected, and more than 700 people ended up in hospital.
Since then, small numbers of cases have popped up from time to time, but without resulting in a large outbreak. There’s currently a number of cases reported in Northland. Measles is highly infectious, and vaccination coverage in New Zealand at the moment is too low to prevent an outbreak.
Using a digital replica of New Zealand, known as ALMA, Fiona and her colleagues ran various simulations of measle outbreaks in different regions and with different starting points. Then they analysed the data to understand more about what might help to slow or stop the spread.
Something they learned from these simulations is that it is still worthwhile to continue vaccinating, even when the disease is already spreading. That’s helpful to plan the best use of resources in a real outbreak, Fiona says.
Fiona Callaghan, chief advisor of epidemiology for the Ministry of Health’s Public Health Agency.
Photo: Supplied
The benefit-harm balance
Beyond applied projects, Andrew Lensen is a director of an AI consultancy, LensenMcGavin, and also researches AI’s social and ethical implications. It’s something he spends a lot of time thinking about, “often at 2am”, he says.
Bias in decision-making, deep fakes, and the impact that American-created AI is having on our culture are all things that keep him awake at night. He and other AI experts recently wrote an open letter to all New Zealand political party leaders, asking for bipartisan support of increased regulation of AI in New Zealand.
For Andrew, this is an important step towards a healthy future for AI in New Zealand. “If we have these conversations about regulation and have these conversations about how it’s used and what is okay in our culture, then [we can] We place the mechanisms in place so that we can use them appropriately, maximize their benefits and minimize their harm. ”
Sign up for a changing world every month Newsletter Episode backstory, scientific analysis, and more.
