How to secure information systems: the growing complexity of the legal landscape

While businesses take a more cautious approach to how they handle IT security threats, these are becoming increasingly complex and sophisticated. Denial-of-service attacks, software tampering (e.g. computer viruses and Trojan horses) and social engineering techniques (e.g. phishing) are some examples becoming prevalent. Developments in IT uses, such as cloud computing, present businesses with the risk of incurring heightened damage from attacks to their IT systems. Consumers are also exposed to the threat of attacks. Orange and Target¹ are currently under scrutiny after having suffered substantial and publicized IT security breaches, potentially exposing their customers to phishing and identity theft. These recent examples of data breaches have cast light on IT security flaws and have led to more focus on the key and impending pieces of legislation that regulate IT security issues in the EU.

Requirement to secure data processing and to notify data breaches

Although the security of data processing is a well-entrenched principle in the EU legal framework, it has recently undergone a significant transformation, notably because of the new notification requirements.

Under Directive 95/46/EC, data controllers have a general duty to implement appropriate technical and organizational measures to protect personal data against various risks including unauthorized disclosure or access.² This obligation also affects data processors. Under French law, non-compliance with such rule constitutes an offence³ and may also trigger an investigation by the CNIL, the French data protection authority, which can render formal and public sanctions, in particular when data controllers fail to restrict access to sensitive information.⁴

Certain data controllers are also required to notify data breaches to the competent national authority, as well as to the data subject under specific conditions.⁵ Although this requirement applies solely to providers of publicly available electronic communications services, namely ISP and telecom operators, it should be extended to all data controllers very soon.⁶ Notification to data subjects can be cumbersome and can damage dramatically the provider’s reputation. However, such an obligation could be waived if the provider demonstrates to the competent national authority that it has taken appropriate technological measures applied to the data concerned to ensure that the data is unintelligible to any person not authorized to access it. However, no specific definition of such technological measures is set forth in Directive 2009/136/EC.

The Commission Regulation of 24 June 2013 provides that encryption with a standardized algorithm and replacement of data by its hashed value are adequate means to render the data unintelligible.⁷ The Regulation also states that implementing such measures are not by themselves sufficient to ensure compliance with the general obligation of security set out in Directive 95/46/EC.⁸ The ability to demonstrate that appropriate technological protection measures have been implemented has been transposed into French law. The CNIL will determine whether the data processor falls under the exemption. Data controllers are only entitled to waive the notification requirement in the case where their practice is documented and is likely to meet the threshold set out by the Regulation.

Requirement to secure networks and information systems

The proposal for a Directive concerning measures to ensure a high common level of network and information security across the Union ("Network and Information Security") includes requirements to take appropriate technical and organizational measures to manage the risks posed to the security of the networks and information systems.⁹ Public administrations, providers of information society services (e.g. social networks, search engines, e-commerce platforms) and operators of critical infrastructure essential for the maintenance of vital economic and societal activities in the fields of energy, transport, banking, stock exchanges and health fall under the scope of the proposal.¹⁰ Incidents having a significant impact on the security of the core services provided by the covered entities will have to be notified to the competent authority. This Proposal is likely to be adopted around mid-2014.

Failure to secure computer facilities may be damaging to businesses

Notwithstanding the obligations mentioned above, businesses are recommended to best protect their information systems, since failure to do so may cause adverse effects in both civil and criminal proceedings. A few recent examples from French case law can be mentioned.

By way of illustration, a French civil court held that a company, seeking damages for unfair competition after one of its employees stole its customer database and transmitted it to a third party, contributed to its own loss by failing to carefully manage the logins.¹¹

From a criminal perspective, the prosecution of wrongdoers who unlawfully access information systems can be hampered by the lack of system protection. Pursuant to the French Criminal Code, anyone who fraudulently accesses or remains in an IT system shall be criminally liable.¹² Although adequate protection of the information system is not a constituent element of the offence, such protection seems to be taken into account by the Courts to assess the fraudulent nature of the access. Recently, the Paris Court of Appeal ruled that fraudulent access had not been established in a case where the access to the system had been enabled by a technical failure attributed to the plaintiff.¹³

The recent Directive 2013/40/EU of 12 August 2013 on attacks against information systems that establishes minimum rules concerning the definition of criminal offences and sanctions in the area of attacks against information systems seems however to require a breach of a security measure to establish the offence of illegal access to information systems.