Public insurer
ACC is reported to be using a secret computer model to predict how long clients will be with the insurer and to pinpoint those it believes are a risk.
The software is said to take advantage of the private information of ACC clients, without their knowledge and possibly without their consent – posing potential mass privacy implications, a report from the
NZ Herald said.
ACC said the algorithm uses information – such as a client’s age or the site of their injury – to make its predictions. However, exactly how the model works is unclear, despite new best practice data guidelines advocating transparency.
Privacy and transparency concerns have swirled since the corporation is viewed to have a poor track record with sensitive information – it was deemed responsible for one of New Zealand’s worst privacy breaches in 2012.
“We need to ask ourselves if this is something that is acceptable to us as New Zealanders and we need to ask how it was allowed to be developed,” specialist ACC lawyer Warren Forster, from advocacy group Acclaim Otago told the
NZ Herald. “We must have transparency and system oversight. People have a right to know what is being done with their data by the government.”
ACC was unable to provide the publication with any information about what ethical guidelines were provided to staff using the model’s predictions for their clients – this is considered a huge concern because ACC is said to be driven by exit targets, the time in which staff can get clients off their books.
“Targets combined with predictions could drive unlawful and unethical behaviour, such as exiting people from the scheme when they still need ACC’s help,” Forster said.
The concern is not about public insurers doing bad things through algorithms, but if they were fair and transparent, Rhema Vaithianathan, co-director of the Centre for Social Data Analytics at AUT told the
NZ Herald.
Related stories:
Drones pose market hazards, says insurer
Former IAG exec joins the ACC board
