Scenario 03: Cyber operation against the power grid

From International cyber law: interactive toolkit
Jump to navigation Jump to search

__NUMBEREDHEADINGS__

© NATO CCD COE

Intelligence services of a State compromise the supply chain of an industrial control system in another State, thereby gaining access to a part of its electric power grid. Subsequent operations bring down the grid, leading to prolonged blackouts. The scenario considers whether such incidents may amount to, among others, a prohibited use of force, an intervention in the internal affairs of another State, or a violation of the sovereignty of another State. Specific consideration is given to whether there exists a standalone obligation to refrain from conducting operations against critical infrastructure of other States through cyber means.

Scenario[edit | edit source]

Keywords[edit | edit source]

Critical infrastructure, prohibition of intervention, self-defence, sovereignty, use of force

Facts[edit | edit source]

[F1] Government-owned company X is responsible for the distribution of electricity across a large part of the territory of State A. Accordingly, its infrastructure has been designated as part of “critical national infrastructure” by the domestic law. Parts of the information infrastructure of company X are used to control critical operations, other parts serve to perform administrative functions unrelated to critical operations.

[F2] Delivery of computers procured as part of the modernisation of the industrial control systems (ICS) used by company X is, unbeknownst to either of the contractual parties, compromised by hackers who succeed in installing concealed remote-control equipment in the computers in question. Once the computers are integrated in the ICS, the hackers are able to remotely monitor the activities in the technical control centre and to assume control over the infrastructure of company X without the staff knowing.

[F3] In the meantime, the relationship between States A and B, frail due to a shared history and a complicated ethnic composition of State A, whom State B periodically accuses of mistreating its large ethnic minority, significantly deteriorates. At one point, the distribution of power to tens of thousands of households in State A suddenly comes to a halt.

[F4] Initially, the technical control centre staff at company X are unable to locate the source of the problem. All reports generated by the ICS suggest normal operation. The controllers are unable to fix the problem remotely and technicians have to be dispatched to individual locations to perform a manual restart and thus to gradually restore the functionality of the network. In the meantime, the government sends its emergency responders and provides generators to the most affected residents.

[F5] As a consequence of the operation, many households are left without electricity for days, resulting in significant inconvenience for the local residents as well as some economic damage to company X and other actors in State A’s territory, including the State itself. However, the power cuts are limited to residential areas and no physical damage or personal injury is reported from any of the affected areas.

[F6] Much later, the source of the vulnerability is identified and the remote-control equipment is found and removed from the ICS at company X. Forensic analysis of the removed equipment determines that it was likely designed, installed, and controlled by the intelligence services of State B. State A is a member of a collective self-defence alliance O.

Examples[edit | edit source]

Legal analysis[edit | edit source]

For a general overview of the structure of analysis in this section, see Note on the structure of articles.

[L1] The analysis in this scenario focusses on the responsibility of State B for potential violations of international law as against State A. It assumes that the cyber operation against company X was attributable to State B. Given the facts of the scenario, this assumption is not particularly controversial. As noted, the technical investigation of the incident showed that the equipment used to compromise the grid had likely been installed by the intelligence service of State B. Pursuant to Article 4 of the ILC Articles on State Responsibility for Internationally Wrongful Acts, the conduct of any State organ, irrespective of its position within the State, its functions and its character as an organ within the central government or territorial unit, shall be considered an act of that State. Intelligence services undoubtedly form part of the executive power and their conduct is thus attributable to the relevant State under Article 4. Accordingly, the remainder of the analysis considers which specific rules of international law, if any, may have been breached by the operation in question.

Use of force[edit | edit source]

{{#lst:Use of force|Definition}} [L2] The scenario notes that the cyber operation against company X had caused significant inconvenience to many households in State A as well as some economic damage to a number of actors including the State itself. However, there is no indication of actual physical damage having occurred or of any injury to individuals as a result of the operation. Therefore, the principal legal question is whether such forms of interference may be categorized as a use of force inconsistent with Article 2(4) of the UN Charter. As noted, it is doubtful whether such interference would amount to “use of force” under contemporary international law.

[L3] In any event, the characterization of an incident of this nature as amounting to a use of force would be of limited consequence in the present scenario. This is because even if a particular act by a State qualifies as prohibited force, the victim State and its allies may only respond in self-defence if the said act is additionally of sufficient gravity to amount to an “armed attack”,<ref> Charter of the United Nations (adopted 26 June 1945, entered into force 24 October 1945) 1 UNTS 16 (UN Charter) Art 51. A minority view should be acknowledged here, according to which the right of self-defense potentially applies against any illegal use of force, irrespective of its qualification as an “armed attack”. See, eg, US DoD, Law of War Manual (December 2016), para 1.11.5.2.</ref> and even then, the permitted response is further limited by the conditions of necessity and proportionality.<ref> See, eg, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v US) (Merits) [1986] ICJ Rep 14, para 194; Legality of the Threat or Use of Nuclear Weapons Case (Advisory Opinion) [1996] ICJ Rep 226, para 41; Oil Platforms (Iran v US) [2003] ICJ Rep 161, para 43.</ref> However, the lack of destructive effects in State A strongly militates against the qualification of the cyber operation by State B as an “armed attack” under international law.<ref> Military and Paramilitary Activities in and against Nicaragua (Nicaragua v US) (Merits) [1986] ICJ Rep 14, para 195 (holding that an operation must be characterized by sufficient “scale and effects” in order to qualify as an “armed attack”); but see Tallinn Manual 2.0, commentary to rule 71, para 12 (noting that some experts held “the view that a cyber operation directed against a State’s critical infrastructure that causes severe, albeit not destructive, effects would qualify as an armed attack“).</ref>

[L4] Destructive effects are reasonably foreseeable and would likely result in loss of life, destruction of property, and significant economic consequences. (e.g. hospitals losing power and patients suffering from a lack of timely treatment.)

[L5] Moreover, the fact that the source of the disruption was only identified after the disruptive effects had been addressed means that at that point, it could no longer be said that a use of force in self-defence by State A or by alliance O was necessary to repel an ongoing attack by State B.<ref> Cf G Nolte and A Randelzhofer, ‘Article 51’ in B Simma et al (eds), The Charter of the United Nations: A Commentary (3rd edn, OUP 2012) vol II, 1426–27, para 60 (noting that the use of force in self-defence is limited to ending the attack so that the specific impulse from which the attack emerged is no longer present); but see David Kretzmer, ‘The Inherent Right to Self-Defence and Proportionality in Jus Ad Bellum’ (2013) 24 EJIL 235, 264–66 (arguing that states that have been the victims of an armed attack may under certain conditions use force to pre-empt future attacks).</ref> Of course, State A would still be entitled to call upon the UN Security Council to qualify the cyber operation as either a “threat to the peace” or a “breach of the peace” and, accordingly, to decide on measures under Chapter VII of the UN Charter.<ref>See Charter of the United Nations (adopted 26 June 1945, entered into force 24 October 1945) 1 UNTS 16 (UN Charter) Art 39.</ref>

[L6] However, the spectrum of unilateral responses available to State A under international law is otherwise identical to those available in response to violations of international law other than the prohibition of force. It is thus arguably unnecessary to conclusively determine if the cyber operation against State A did in fact cross the threshold of Article 2(4) of the UN Charter, as long as the relevant conduct breached other applicable international legal rules. This is what the remainder of the analysis turns to.

Prohibition of intervention[edit | edit source]

{{#lst:Prohibition of intervention|Definition}} [L7] It is unclear whether State B’s act had a bearing on State A’s domaine réservé in the present case (the first prong of the test). State A would have to establish that “the act in question [was] designed to undermine [its] authority over the domaine réservé”.<ref> Tallinn Manual 2.0, commentary to rule 66, para 11.</ref> If, for instance, it transpired that State B designed the operation with the specific aim to bring to an end the ethnic discrimination by organs of State A or to affect an aspect of State A’s foreign policy, then this would bring it within the protected sphere of matters, fulfilling the first prong of the test.

[L8] As for the element of coercion, it depends, as above, on the purpose of the operation. If it was designed to compel State A to modify its policing practice, then (whether or not this outcome was in fact achieved) the second prong of the test would have been fulfilled, too.<ref> Tallinn Manual 2.0, commentary to rule 66, para. 29 (“the fact that a coercive cyber operation fails to produce the desired outcome has no bearing on whether [the prohibition of intervention] has been breached”).</ref>

Obligation to respect the sovereignty of other States[edit | edit source]

{{#lst:Sovereignty|Definition}} [L9] Whether sovereignty has been violated in the present case is controversial. The outcome of the operation was limited to physically non-destructive effects and its impact on the electrical distribution grid was fully reversible, and therefore whether this amounts to a violation of territorial sovereignty depends inter alia on whether one considers that a State’s right to territorial sovereignty extends to intangible layers of cyberspace.

[L10] For some, emphasis should be placed on the fact that the operation caused significant inconvenience for the population and necessitated a degree of government response in the form of provision of supplies and aid. On that basis, they would submit that operations of this kind qualify as infringing on the sovereignty of the affected State given that the need to respond deprives that State of a part of its ability to conduct its affairs freely.

Possible obligation not to conduct cyber operations against other States’ critical infrastructure[edit | edit source]

[L11] In its 2015 report, the UN group of governmental experts agreed on a formulation that “[a] State should not conduct or knowingly support ICT activity contrary to its obligations under international law that intentionally damages critical infrastructure or otherwise impairs the use and operation of critical infrastructure to provide services to the public“.<ref> UN GGE 2015 'Report of the Group of Governmental Experts on Developments in the Field of Information and Telecommunications in the Context of International Security report' (22 July 2015) UN Doc A/70/174, para. 13(f).</ref> This raises the question whether a cyber operation such as the one described in this scenario infringes an obligation not to conduct operations against the critical national infrastructure of other States.

[L12] However, it is doubtful whether such an obligation can be said to exist in the present state of international law. Three points should be made in this regard. First of all, cyber operations against critical infrastructure have recently become a very frequent occurrence in the international practice.<ref> See, eg, US, Director of National Intelligence James Clapper, Worldwide Threat Assessment of the US Intelligence Community (26 February 2015) (“foreign actors are reconnoitering and developing access to U.S. critical infrastructure systems, which might be quickly exploited for disruption if an adversary’s intent became hostile.”); FO Hampson and M Sulmeyer (eds), Getting Beyond Norms (CIGI 2017) 6 (“Disrupting or damaging critical infrastructures that provide services to the public has become customary practice — the new normal. In the past two years and since the GGE agreement, there have been an alarming number of harmful incidents targeting critical infrastructures around the world, ranging from power systems to telecommunications systems to transportation systems to financial systems.”).</ref> Of course, the frequency of a particular type of conduct, even if it “amount[s] to a settled practice”, does not by itself suffice to establish a new permissive rule of customary law.<ref> Military and Paramilitary Activities in and against Nicaragua (Nicaragua v US) (Merits) [1986] ICJ Rep 14, para 207, citing ICJ, North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands) [1969] ICJ Rep 3, para 77.</ref> However, the fact that most of these instances have not been accompanied by specific condemnations by those States in a position to react to them, suggests that equally, no corresponding prohibitive rule has emerged thus far.<ref> Cf. also ECCC, Case No 002/19-09-2007-EEEC/ OICJ (PTC38), Decision on the Appeals Against the Co-Investigative Judges Order on Joint Criminal Enterprise (JCE) (20 May 2010), para 53 (“A wealth of State practice does not usually carry with it a presumption that opinio juris exists”).</ref>

[L13] Secondly, although UN GGE reports are based on the consensus of governmental experts selected on the basis of equitable geographical representation,<ref> UNGA Res 68/243 (27 December 2013) UN Doc A/RES/68/243, para 4.</ref> their views do not necessarily reflect customary international law, nor should they be seen as aiming to do so. The mandate of the 2014–15 GGE was in any case not focussed on the identification of customary legal rules; rather, it included the proposing and promoting of “[v]oluntary, non-binding norms of responsible State behaviour”.<ref> UN GGE 2015 'Report of the Group of Governmental Experts on Developments in the Field of Information and Telecommunications in the Context of International Security report' (22 July 2015) UN Doc A/70/174, para 10. On the distinction between cyber rules and cyber norms, see K Mačák, ‘From Cyber Norms to Cyber Rules: Re-engaging States as Law-makers’ (2017) 30 LJIL 877, 877–99.</ref> Given that the experts specifically described the formulation quoted in para. L11 above as one of several “recommendations for consideration by States”,<ref>UN GGE 2015 'Report of the Group of Governmental Experts on Developments in the Field of Information and Telecommunications in the Context of International Security report' (22 July 2015) UN Doc A/70/174, para 13.</ref> it should not be understood as purporting to identify a new rule of custom.

[L14] Thirdly, it is manifest from the formulation itself that the governmental experts did not consider that any intentional cyber operation against critical national infrastructure would automatically be internationally unlawful. This is confirmed by the use of the word “should”, which is generally considered hortatory, as opposed to “shall” or “must”.<ref>See, eg, Robert Rosenstock, ‘The Declaration of Principles of International Law concerning Friendly Relations: A Survey’ (1971) 65 AJIL 713, 715; Pierre-Marie Dupuy, ‘Soft Law and the International Law of the Environment’ (1991) 12 Michigan J Intl L 420, 429.</ref> Moreover, the experts included the phrase “contrary to its obligations under international law”, confirming that on their view, an operation against critical infrastructure abroad would in any event only violate international law if it breached a separate international obligation.<ref>Cf. UN GGE 2015 'Report of the Group of Governmental Experts on Developments in the Field of Information and Telecommunications in the Context of International Security report' (22 July 2015) UN Doc A/70/174, para. 13(f).</ref>

[L15] In sum, the preferred view is that a standalone rule prohibiting cyber operations against critical national infrastructure has not emerged in international law thus far. As such, the incident in the scenario cannot be described as infringing this supposed obligation.

Checklist[edit | edit source]

  • Use of force:
    • Did the operation result in actual physical damage or injury to individuals?
  • Prohibition of intervention:
    • Did the operation bear on any of those matters in which States are allowed to decide freely?
    • Did the operation amount to a coercive act against the victim State?
  • Sovereignty:
    • What is the position of the client on whether sovereignty is a standalone primary rule of international law?
    • Were any individuals associated with an outside State physically present in the domestic State’s territory without the latter’s consent?
    • Did the operation occasion a loss of functionality of cyber infrastructure?
    • Did the operation interfere with or usurp inherently governmental functions of another State?
  • Critical infrastructure:
    • What is the position of the client on whether there is a standalone rule prohibiting cyber operations against critical national infrastructure?

Appendixes[edit | edit source]

See also[edit | edit source]

Notes and references[edit | edit source]

<references/>

Bibliography and further reading[edit | edit source]

Contributions[edit | edit source]

Previous: Scenario 02: Political espionage Next: Scenario 04: International organizations