In the fifth episode of Actually Interesting, The Spinoff’s monthly podcast exploring the effect Artificial Intelligence has on our lives, Russell Brown speaks to Ben Reid, executive director of the AI Forum, about the role of government in embracing and regulating AI.
Here’s a thing you probably didn’t know: if you’ve sought ACC cover in the past couple of years, your claim was very likely processed by artificial intelligence.
More precisely, your likelihood of cover was assessed by an algorithm trained on an anonymised dataset of 12 million claims made between 2010 and 2016. The AI system embodies a predictive model that decides whether or not your claim clearly falls within the criteria of the Accident Compensation Act 2001. Most claims do, and are thus swiftly and efficiently approved. Uncertain cases or potential refusals are then handled by human staff. It’s designed so the computer literally can not say “no”.
ACC’s system is one of a number of case studies cited in Towards Our Intelligent Future: An AI Roadmap for New Zealand, the new report from the AI Forum New Zealand. The report is a substantial work – 180 pages dedicated to explanations of the key AI technologies, a look at the international AI landscape (which is dominated by two research and investment superpowers, the US and China) – and a polite, repeated request for a national AI strategy.
The forum made essentially the same call in a report last year and, in a landscape of task forces and working groups, we still haven’t seen that strategy. Is government listening? Ben Reid, the AI Forum’s executive director, is both diplomatic and optimistic.
“They’re beginning to. Just recently a group of us from the AI Forum presented to the select committee on economic development, science and innovation and that was the first time I think that AI’s been considered by Parliament. I think it’s a really positive sign.”
The forum’s decision to downplay the technology in the section of the report that offers an AI roadmap for New Zealand – talking instead about outcomes, wellbeing and sustainability – is a clear bid by the forum to make its case in the current government’s language.
“It’s totally deliberate,” says Reid. “If we talk about general understanding, people aren’t always aware that AI technologies are actually around us ubiquitously today. For example, when you’re online shopping and you get a recommendation, you’re shown that other people who bought this, bought another item.
“People use these consumer services, but don’t see how AI can extend to New Zealand’s highest-priority objectives. When you think about the number of people that are dying on our roads, the number dying from cancer, the challenges we have around conservation of the environment. And with the government’s focus on wellbeing really setting objectives, what we’ve tried to do in this report is to show how in a tangible way AI can be lined up against New Zealand’s objectives.”
Let’s go back to ACC. The key asset in its AI cover approval system is its database of claims. The bigger the set, the better the AI’s ability to recognise patterns. We’re tracking towards a day when huge quantities of data are generated by the sector that has often driven tech development in New Zealand – agriculture. There’s a precision farming revolution quietly underway, much of it using Internet of Things technologies, and it’s going to produce a mountain of data.
“One of the near-term opportunities is driven by the sheer amount of aerial and satellite data that’s starting to be produced around the world,” Reid agrees. “There’s a fantastic company out of Christchurch called Orbica, who are specialists in what they call GeoAI – analysing geospatial data and aerial and satellite photography to extract features and effectively count the number of houses or vehicles or identify bodies of water from the air. These are things that used to take people days of manual analysis and now it can be run through an algorithm in seconds.
“You can run an analysis from a drone flying across your vineyard for example and get a prediction on the yield that you’re going to get this year.”
Reid believes that capturing and sharing that data within and between productive sectors could represent a vital economic advantage.
“One of the concepts we talk about in the report is that of a data trust. This is where commercially sensitive data from a number of stakeholders can be pooled under a legal structure called a data trust which enables collective outcomes from that data.
“An example of that is the tourism sector. Often they’re being undercut by international competition, the Air B&Bs, or booking.com, who’ve got to get better models of where tourism demand is going to be. Could the New Zealand tourism sector create Data of National Significance in a way that enables tourism operators here, maybe similar to a Fonterra or a Zespri in some ways, to manage that data across the tourism sector and give New Zealand tourism operators a commercial advantage?”
The forum also wants the government to create a regulatory framework that recognises that not all AI activity is benign or beneficial. Cambridge Analytica, after all, was able to do what it did because it illicitly harvested 50 million Facebook users’ personal information.
“Obviously we want on a national level to have regulation that’s consistent with regulations happening around the world – for example, the GDPR out of Europe has become a de facto standard internationally. How do policy-makers understand the technology in order to make good policy decisions? That’s an area that we’ve really focused on and we’ve created a policy-map which shows across lots of portfolios in government how AI may intersect with some of the decisions.”
And yet, the GDPR isn’t the only thing going on. China, one of those two developing AI superpowers, is working its way towards a surveillance state. Reid is still hopeful that there will be a global way forward.
“There’s almost AI diplomacy being done. So personally I’m optimistic you’ll see the two traditions work out their differences and come up with a framework that works around the world.”
And where does New Zealand fit in with the ethics of giants?
“We need to be actively involved, in our multilateral tradition.”
AI Forum New Zealand is a group representing New Zealand’s community of artificial intelligence technology innovators, end-users, investor groups, regulators, researchers, educators, entrepreneurs and interested members of the public.
This content was created in paid partnership with Microsoft. Learn more about our partnerships here.
Subscribe to The Bulletin to get all the day’s key news stories in five minutes – delivered every weekday at 7.30am.