Under Australian privacy laws, organisations need to implement ‘reasonable steps’ to protect any personal information they hold against misuse, interference, loss, unauthorised access, modification or disclosure. We look at the Privacy Commissioner’s determinations for guidance on what should be done.
Recently, when Australia’s privacy regulator (the OAIC) was asked about its privacy governance priorities for 2016, the issue of information security was, unsurprisingly, high on the agenda.
The OAIC emphasised the need for corporate counsel to focus on “get(ting) security right and…mak(ing) sure that there is a real culture of building in privacy and security protections in organisations.”
Growing concerns around cyber security and the prospect of more stringent regulation of data breaches have prompted many businesses to reassess their information security practices.
While the Privacy Commissioner has issued best practice guidelines on applying the ‘reasonable steps’ requirement, these guidelines do not provide a ‘bright line’ test for minimum compliance. This has led to some uncertainty as to how the ‘reasonableness’ standard will be applied in the context of today’s rapidly evolving cyber security landscape.
HOW TO DETERMINE ‘REASONABLE STEPS’: 5 KEY LESSONS
Below, we’ve set out five key lessons that can be extracted from the Commissioner’s published reports on data breach investigations, in the hope they will assist in framing what ‘reasonable steps’ are.
LESSON 1: BE AWARE THAT ‘REASONABLE STEPS’ IS NOT THE SAME AS ‘REASONABLE INDUSTRY PRACTICE’
It is clear from the Commissioner’s investigations that there is no ‘tick the box’ approach to regulatory compliance. This is because the Commissioner has:
rejected arguments that ‘reasonableness’ should be equated with ‘industry practice’ or what is ‘commercially’ reasonable; and
found that engaging external providers to advise on and manage information security may not, in and of itself, be sufficient to discharge the “reasonable steps” requirement.
While the above factors may be relevant to assessing ‘reasonable steps’, organisations should not rely exclusively on these external benchmarks as a substitute for giving specific consideration to their own risk profile (based on factors such as the quantity and sensitivity of the organisations’ personal information holdings).
LESSON 2: ALWAYS ACCEPT OWNERSHIP OF THE RISK
A clear message from the investigation reports is that organisations have a duty to independently consider their particular needs and circumstances and implement appropriate information security measures.
Your organisation must take ownership of its self-assessment, and cannot outsource ownership of the risk to external consultants or advisers. While organisations should seek expert advice where appropriate, final responsibility for cyber security remains with your organisation.
In this regard, it is helpful to manage information security as a corporate governance issue – that is, organisations should view information security as an enterprise-wide business risk that requires continual monitoring, effective governance and appropriate oversight from the Board and senior management.
LESSON 3: MANAGE COSTS BY ADOPTING A SYSTEMATIC, RISK-BASED APPROACH
Cost is obviously a key consideration when developing an information security strategy.
The Commissioner has made it clear that an organisation will only be excused from implementing a security measure that would otherwise be considered “reasonable” if the cost or inconvenience of implementation would be “excessive”.
An effective way to manage costs is to adopt a risk-based approach to information security management. This is also consistent with regulatory guidance from the Commissioner and ASIC, who both recommend that organisations should structure their security programs around identifying and addressing key areas of risk.
Implementing a recognised information security standard such as the NIST framework (which has been recommended by ASIC) and the ISO/IEC 27000 series can help your organisation take a systematic, risk-based approach to managing information security risk. These standards are generally scalable in nature, and can be implemented in a way that fits the particular risk profile and resources of your organisation.
Further, implementing a recognised standard does not guarantee compliance with the Privacy Act’s security requirements, but it certainly helps. The Commissioner has referred approvingly to standards such as the ISO/IEC 27000 series in both his best practice guidelines, and in certain investigation reports.
LESSON 4: YOUR SECURITY IS ONLY AS GOOD AS ITS WEAKEST LINK
It is critical for businesses to ensure that adequate security measures are implemented consistently across all business systems and processes.
In the event of a security breach investigation, the Commissioner will assess whether the business had implemented “reasonable” security in relation to the particular system or process that was actually exploited in the breach (rather than considering whether the business’ overall security framework was “reasonable” as a whole).
Therefore, organisations should ensure that security testing procedures are suitably comprehensive, and go beyond the obvious ‘high value’ IT assets. Organisations need to be vigilant to ‘weak points’ that might not be immediately obvious.
LESSON 5: BE AWARE OF ISSUES PREVIOUSLY CONSIDERED BY THE COMMISSIONER
The Commissioner generally avoids making definitive statements about information security standards (for example, encryption standards) in his investigation reports. He is careful to emphasise that there is no ‘one size fits all’ approach, and that the particular circumstances of each case must be considered on its own merits.
Nevertheless, organisations should be aware of the types of security practices that the Commissioner has previously held to be insufficient in his investigation reports. These provide a useful benchmark for organisations to measure their security practices against.
Some recent examples of security practices that the Commissioner has found to be insufficient (and should be avoided) include:
- Storing user e-mails, passwords and password hints in plain text, or only encrypting them with a single encryption key and ‘block cipher’ encryption algorithm.
- Storing user data on a publically-accessible web server.
- Failing to identify a system vulnerability due to limited, sample-based security testing.
- Issuing consumer equipment to customers without conducting adequate testing to determine whether factory default settings provided appropriate access controls and privacy protections for the customer.
- Failing to conduct appropriate software maintenance and lifecycle management (e.g. patch deployment).
- Using cryptographic protocols such as SSL to secure data in transit, without ensuring that the destination location is also secure.
- Failing to properly configure a website to disable directory browsing and content indexation by search robots (such as Googlebot).
- Failing to promptly purge old user data once it becomes out of date, and/or failing to promptly decommission old servers.
- Failing to have appropriate policies regarding the redaction of personal information from published reports, and failing to properly educate staff on the importance of complying with those policies.
- Storing hard copy documents containing personal information on premises that are not adequately secured or monitored by the organisation.
The five key lessons extracted from the Privacy Commissioner’s determinations and outlined in this article should assist organisations to determine what ‘reasonable steps’ are in relation to the protection of personal information. They are not a definitive set of rules though and need to be amended against the particular systems in place.