Optus suffered a major data breach in 2022. Detail from Bianca De Marchi/AAP^[7]

They have to notify both the regulator – the Office of the Australian Information Commissioner (OAIC) – and any affected individuals of breaches that are likely to result in “serious harm^[8]”.

But according to the organisational leaders we interviewed, this poses a tricky question. How do you define serious harm?

Interpretations of what “serious harm” actually means – and how likely it is to occur – vary significantly. This inconsistency can make it impossible to predict the specific impact of a data breach on an individual.

Victims of domestic violence, for example, may be at increased risk when personal information is exposed, creating harms that are difficult to foresee or mitigate.

Enforcing the rules

Interviewees also had concerns about how well the regulator could provide guidance and enforce data protection measures.

Many expressed a belief the OAIC is underfunded and lacks the authority to impose and enforce fines properly. The consensus was that the challenge of protecting our data has now outgrown the power and resources of the regulator.

As one chief information security officer at a publicly listed company put it:

What’s the point of having speeding signs and cameras if you don’t give anyone a ticket?

A lack of enforcement can undermine the incentive for organisations to invest in robust data protection.

Only the tip of the iceberg

Data breaches are also underreported, particularly in the corporate sector.

One senior cybersecurity consultant from a major multinational company told us there is a strong incentive for companies to minimise or cover up breaches, to avoid embarrassment.

This culture means many breaches that should be reported simply aren’t. One senior public servant estimated only about 10% of reportable breaches end up actually being disclosed.

Without this basic transparency, the regulator and affected individuals can’t take necessary steps to protect themselves.

Affected individuals can’t take steps to protect themselves if breaches aren’t reported. Yuri A/Shutterstock^[9]

Third-party breaches

Sometimes, when we give our personal information to one organisation, it can end up in the hands of another one we might not expect. This is because key tasks – especially managing databases – are often outsourced to third parties.

Outsourcing tasks might be a more efficient option for an organisation, but it can make protecting personal data even more complicated.

Interviewees told us breaches were more likely when engaging third-party providers, because it limited the control they had over security measures.

Between July and December 2023 in Australia, there was an increase of more than 300%^[10] in third-party data breaches compared to the six months prior.

There have been some highly publicised examples.

In May this year, many Clubs NSW customers had their personal information potentially breached^[11] through an attack on third-party software provider Outabox.

Bunnings suffered a similar breach^[12] in late 2021, via an attack on scheduling software provider FlexBooker.

In 2021, Bunnings had outsourced some customer booking tasks to third-party provider Flexabooker. Dave Hunt/AAP^[13]

Getting the basics right

Some organisations are still struggling with the basics. Our research found many data breaches occur because outdated or “legacy” data systems are still in use.

These systems are old or inactive databases, often containing huge amounts of personal information about all the individuals who’ve previously interacted with them.

Organisations tend to hold onto personal data longer than is legally required. This can come down to confusion about data-retention requirements, but also the high cost and complexity of safely decommissioning old systems.

One chief privacy officer of a large financial services institution told us:

In an organisation like ours where we have over 2,000 legacy systems […] the systems don’t speak to each other. They don’t come with big red delete buttons.

Other interviewees flagged that risky data testing practices are widespread.

Software developers and tech teams often use “production data” – real customer data – to test new products. This is often quicker and cheaper than creating test datasets.

However, this practice exposes real customer information to insecure testing environments, making it more vulnerable. A senior cybersecurity specialist told us:

I’ve seen it so much in every industry […] It’s literally live, real information going into systems that are not live and real and have low security.

What needs to be done?

Drawing insights from professionals at the coalface, our study highlights just how complex data protection has become in Australia, and how quickly the landscape is evolving.

Addressing these issues will require a multi-pronged approach, including clearer legislative guidelines, better enforcement, greater transparency and robust security practices for the use of third-party providers.

As the digital world continues to evolve, so too must our strategies for protecting ourselves and our data.

Read more https://theconversation.com/why-do-organisations-still-struggle-to-protect-our-data-we-asked-50-professionals-on-the-privacy-front-line-236681