Standardisation of Cybersecurity for Artificial Intelligence

The European Union Agency for Cybersecurity (ENISA) publishes an assessment of standards for the cybersecurity of AI and issues recommendations to support the implementation of upcoming EU policies on Artificial Intelligence (AI).

This report focuses on the cybersecurity aspects of AI, which are integral to the European legal framework regulating AI, proposed by the European Commission last year dubbed as the “AI Act“.

What is Artificial Intelligence?

The draft AI Act provides a definition of an AI system as “software developed with one or more (…) techniques (…) for a given set of human-defined objectives, that generates outputs such as content, predictions, recommendations, or decisions influencing the environments they interact with.” In a nutshell, these techniques mainly include: machine learning resorting to methods such as deep learning, logic, knowledge-based and statistical approaches.

It is indeed essential for the allocation of legal responsibilities under a future AI framework to agree on what falls into the definition of an 'AI system'.

However, the exact scope of an AI system is constantly evolving both in the legislative debate on the draft AI Act, as well in the scientific and standardisation communities.

Although broad in contents, this report focuses on machine learning (ML) due to its extensive use across AI deployments. ML has come under scrutiny with respect to vulnerabilities particularly impacting the cybersecurity of an AI implementation.

AI cybersecurity standards: what’s the state of play?

As standards help mitigate risks, this study unveils existing general-purpose standards that are readily available for information security and quality management in the context of AI. In order to mitigate some of the cybersecurity risks affecting AI systems, further guidance could be developed to help the user community benefit from the existing standards on AI.

This suggestion has been based on the observation concerning the software layer of AI. It follows that what is applicable to software could be applicable to AI. However, it does not mean the work ends here. Other aspects still need to be considered, such as:

  • a system-specific analysis to cater for security requirements deriving from the domain of application;
  • standards to cover aspects specific to AI, such as the traceability of data and testing procedures.

Further observations concern the extent to which the assessment of compliance with security requirements can be based on AI-specific horizontal standards; furthermore, the extent to which this assessment can be based on vertical/sector specific standards calls for attention.

Key recommendations include:

  • Resorting to a standardised AI terminology for cybersecurity;
  • Developing technical guidance on how existing standards related to the cybersecurity of software should be applied to AI;
  • Reflecting on the inherent features of ML in AI. Risk mitigation in particular should be considered by associating hardware/software components to AI; reliable metrics; and testing procedures;
  • Promoting the cooperation and coordination across standards organisations’ technical committees on cybersecurity and AI so that potential cybersecurity concerns (e.g., on trustworthiness characteristics and data quality) can be addressed in a coherent manner.

Regulating AI: what is needed?

As for many other pieces of EU legislation, compliance with the draft AI Act will be supported by standards. When it comes to compliance with the cybersecurity requirements set by the draft AI Act, additional aspects have been identified. For example, standards for conformity assessment, in particular related to tools and competences, may need to be further developed. Also, the interplay across different legislative initiatives needs to be further reflected in standardisation activities – an example of this is the proposal for a regulation on horizontal cybersecurity requirements for products with digital elements, referred to as the “Cyber Resilience Act”.

Building on the report and other desk research as well as input received from experts, ENISA is currently examining the need for and the feasibility of an EU cybersecurity certification scheme on AI. ENISA is therefore engaging with a broad range of stakeholders including industry, ESOs and Member States, for the purpose of collecting data on AI cybersecurity requirements, data security in relation to AI, AI risk management and conformity assessment.

ENISA advocated the importance of standardisation in cybersecurity today, at the RSA Conference in San Francisco in the ‘Standards on the Horizon: What Matters Most?’ in a panel comprising the National Institute of Standards and Technology (NIST).

CISA and Partners Disclose Snake Malware Threat From Russian Cyber Actors

CISA and partners released a joint advisory for a sophisticated cyber espionage tool used by Russian cyber actors. Hunting Russian Intelligence “Snake” Malware provides technical descriptions of the malware’s host architecture and network communications, and mitigations to help detect and defend against this threat.

 

The Snake implant is considered the most sophisticated cyber espionage tool designed and used by Center 16 of Russia’s Federal Security Service (FSB) for long-term intelligence collection on sensitive targets. To conduct operations using this tool, the FSB created a covert peer-to-peer (P2P) network of numerous Snake-infected computers worldwide. Many systems in this P2P network serve as relay nodes which route disguised operational traffic to and from Snake implants on the FSB’s ultimate targets. Snake’s custom communications protocols employ encryption and fragmentation for confidentiality and are designed to hamper detection and collection efforts.

CISA has identified Snake infrastructure in over 50 countries across North America, South America, Europe, Africa, Asia, and Australia, to include the United States and Russia itself. Although Snake uses infrastructure across all industries, its targeting is purposeful and tactical in nature. Globally, the FSB has used Snake to collect sensitive intelligence from high-priority targets, such as government networks, research facilities, and journalists. As one example, FSB actors used Snake to access and exfiltrate sensitive international relations documents, as well as other diplomatic communications, from a victim in a North Atlantic Treaty Organization (NATO) country. Within the United States, the FSB has victimized industries including education, small businesses, and media organizations, as well as critical infrastructure sectors including government facilities, financial services, critical manufacturing, and communications.

This Cybersecurity Advisory (CSA) provides background on Snake’s attribution to the FSB and detailed technical descriptions of the implant’s host architecture and network communications. This CSA also addresses a recent Snake variant that has not yet been widely disclosed. The technical information and mitigation recommendations in this CSA are provided to assist network defenders in detecting Snake and associated activity. For more information on FSB and Russian state-sponsored cyber activity, please see the joint advisory Russian State-Sponsored and Criminal Cyber Threats to Critical Infrastructure and CISA’s Russia Cyber Threat Overview and Advisories webpage.

CISA urges organizations to review the advisory for more information and apply the recommended mitigations and detection guidance.

The One Sea Association and ESA partner to support the uptake of autonomous shipping in the maritime sector

The One Sea Association and the European Space Agency (ESA) have signed a Memorandum of Intent (MoI) to support the uptake of autonomous shipping in the maritime sector, underpinned by space solutions.

One Sea and ESA have decided to establish a strategic collaboration to promote the development of new space-enabled services which will support the maritime sector’s transition towards autonomous shipping. Autonomous shipping offers new opportunities to deploy safe, commercially viable, and environmentally sustainable maritime operations.

Satellite communications and satellite navigation play a key role in the adoption of autonomous shipping technologies and operations. During offshore passages, ships are often further from land than satellites which can offer invaluable secure and resilient communication channels for monitoring, command, and control of autonomous ships. Furthermore, in ports and congested areas, high precision Position Navigation and Timing (PNT) provided by satellites is also critical for the safe operation of autonomous shipping.

This new partnership will combine One Sea’s unique expertise in the maritime sector and in autonomous shipping with ESA’s technical competence and mandate through the Business Applications and Space Solutions programme to support the development and demonstration of space solutions in addressing user needs.

 

CISA Urges Organizations to Incorporate the FCC Covered List Into Risk Management Plans

The Federal Communications Commission (FCC) maintains a Covered List of communications equipment and services that have been determined by the U.S. government to pose an unacceptable risk to the national security of the United States or the security and safety of United States persons to national security pursuant to the Secure and Trusted Communications Networks Act of 2019.

As the 6th annual National Supply Chain Integrity Month concludes, CISA reminds all critical infrastructure owners and operators to take necessary steps in securing the nation’s most critical supply chains. CISA urges organizations to incorporate the Covered List into their supply chain risk management efforts, in addition to adopting recommendations listed in Defending Against Software Supply Chain Attacks—a joint CISA and NIST resource that provides guidance on using NIST’s Cyber Supply Chain Risk Management (C-SCRM) framework to identify, assess, and mitigate risks. All critical infrastructure organizations are also urged to enroll in CISA’s free Vulnerability Scanning service for assistance in identifying vulnerable or otherwise high-risk devices such as those on FCC’s Covered List.

To learn more about CISA’s supply chain efforts and to view resources, visit CISA.gov/supply-chain-integrity-month.

NERC files report evaluating the CIP-014 Reliability Standard with FERC

The Commission directed NERC to evaluate whether the physical security protection requirements in NERC’s Reliability Standards are adequate to address the risks associated with physical attacks on BPS Facilities. Specifically, FERC directed NERC to conduct a study evaluating the following: (1) the adequacy of the Applicability criteria set forth in the Physical Security Reliability Standard; (2) the adequacy of the required risk assessment set forth in the Physical Security Reliability Standard; and (3) whether a minimum level of physical security protections should be required for all BPS substations and their associated primary control centers.

The purpose of the CIP-014 Reliability Standard is to “identify and protect Transmission stations and Transmission substations, and their associated primary control centers, that if rendered inoperable or damaged as a result of a physical attack could result in instability, uncontrolled separation, or Cascading within an Interconnection.”2 The standard requires applicable Transmission Owners (“TOs”) to perform periodic risk assessments of their applicable transmission stations and transmission substations (hereinafter collectively referred to as “substations”) to identify which of their applicable substations are “critical” to BPS reliability (which, for purposes of CIP-014, is whether instability, uncontrolled separation, or Cascading would result if the substation were damaged or rendered inoperable). The TO must then perform an evaluation of the potential physical security threats and vulnerabilities of a physical attack to each of their “critical” substations and develop and implement a documented physical security plan to address those threats and vulnerabilities. Additionally, for each primary control center that operationally controls an identified substation, the applicable Transmission Operator (“TOP”) must perform an evaluation of the potential physical security threats and vulnerabilities of a physical attack to that control center and develop and implement a documented physical security plan to address those threats and vulnerabilities.

As discussed within this report, NERC finds that the objective of CIP-014 appropriately focuses limited industry resources on risks to the reliable operation of the BPS associated with physical security incidents at the most critical facilities. Based on studies using available data, NERC finds that the CIP-014 Applicability criteria is meeting that objective and is broad enough to capture the subset of applicable facilities that TOs should identify as “critical” pursuant to the risks assessment mandated by Requirement R1. NERC did not find evidence that an expansion of the Applicability criteria would identify additional substations that would qualify as “critical” substations under the CIP- 014 Requirement R1 risk assessment. Accordingly, at this time, NERC is not recommending expansion of the CIP-014 Applicability criteria.

NERC acknowledges, however, that supplementary data3 could show that additional substation configurations would warrant assessment under CIP-014. Accordingly, NERC plans to continue evaluating the adequacy of the Applicability criteria in meeting the objective of CIP-014. Following issuance of this report, NERC will work with FERC staff to hold a technical conference to, among other things, identify the type of substation configurations that should be studied to determine whether any additional substations should be included in the Applicability criteria. The technical conference would also help establish data needs for conducting those studies

NERC finds, however, that the language in Requirement R1 of CIP-014 should be refined to ensure that entities conduct effective risk assessments of their applicable substations. Information from ERO Enterprise Compliance Monitoring and Enforcement Program (“CMEP”) activities indicates that while the overall objective of the CIP-014 Requirement R1 risk assessment is sound, there are inconsistent approaches to performing the risk assessment. The ERO Enterprise observed that, in certain instances, registered entities failed to provide sufficient technical studies or justification for study decisions resulting in noncompliance. NERC finds that the inconsistent approach to performing the risk assessment is largely due to a lack of specificity in the requirement language as to the nature and parameters of the risk assessment. Accordingly, NERC will initiate a Reliability Standards development project to evaluate changes to CIP-014 to provide additional clarity on the risk assessment.

As discussed further below, the objective of the Reliability Standards development project would be to:
• Clarify the risk assessment methods for studying instability, uncontrolled separation, and Cascading; such as the expectations of dynamic studies to evaluate for instability.
• Clarify the case(s) used for the assessment to be tailored to the Requirement R1 in-service window and correct any discrepancies between the study period, frequency of study, and the base case a TO uses.
• Clarify the documentation, posting, and usage of known criteria to identify instability, uncontrolled separation, or Cascading as part of the risk assessment. The criteria should also include defining “inoperable” or “damaged” substations such that the intent of the risk assessment is clear.
• Clarify the risk assessment to account for adjacent substations of differing ownership, and substations within line-of-sight to each other.

Finally, while NERC is not recommending an expansion of the CIP-014 Applicability criteria at this time, NERC finds that, given the increase in physical security attacks on BPS substations, there is a need to evaluate additional reliability, resiliency, and security measures designed to mitigate the risks associated with those physical security attacks. As discussed further below, establishing a uniform, bright line set of minimum physical security protections for all (or even an additional subset of) BPS substations and associated primary controls centers, is unlikely to be an effective approach to mitigating physical security risks and their potential impacts on the reliable operation of the BPS. While a uniform set of minimum level of protections could potentially prevent some forms of physical security threats, NERC finds that such a pursuit lacks the application of a risk-based approach to expending industry resources, fails to provide for a methodical approach necessary to address site-specific threats or objectives (as expected using a design basis threat process), and does not consider the need for other reliability, resiliency, and security measures to mitigate the impact of a physical attack. These combined measures provide increased operational and planning capability as well as improved effectiveness of local network restoration. NERC finds that this more holistic approach will provide greater long-term flexibility and minimize the impacts of physical attacks on BPS reliability.

 

Full report can be found here >>

Time Frames to Complete CISA Efforts Would Help Sector Risk Management Agencies Implement Statutory Responsibilities

Protecting critical infrastructure that helps provide necessities like water, electricity, and food is a national priority. Events like natural disasters or cyberattacks can disrupt services Americans need for daily life.

We testified that many federal agencies work to protect the nation's critical infrastructure and look to the Cybersecurity and Infrastructure Security Agency for leadership on how to do it.

A 2021 law expanded these agencies' responsibilities and added some new ones. CISA is working on guidance and more to help agencies implement these responsibilities. We've recommended that CISA set timelines for completing this work.

The William M. (Mac) Thornberry National Defense Authorization Act for Fiscal Year 2021 expanded and added responsibilities for Sector Risk Management Agencies (SRMAs). These agencies engage with their public and private sector partners to promote security and resilience within their designated critical infrastructure sectors. Some officials from these agencies described to GAO new activities to address the responsibilities set forth in the act, and many reported having already conducted related activities. For example, the act added risk assessment and emergency preparedness as responsibilities not previously included in a key directive for SRMAs. New activities officials described to address these responsibilities included developing a communications risk register and developing emergency preparedness exercises.

The Department of Homeland Security's (DHS) Cybersecurity and Infrastructure Security Agency (CISA) has identified and undertaken efforts to help SRMAs implement their statutory responsibilities. For example, CISA officials stated they are updating key guidance documents, including the 2013 National Infrastructure Protection Plan and templates for revising sector-specific guidance documents. CISA officials also described efforts underway to improve coordination with sector partners, such as reconvening a leadership council. SRMA officials for a majority of critical infrastructure sectors reported that additional guidance and improved coordination from CISA would help them implement their statutory responsibilities. However, CISA has not developed milestones and timelines to complete its efforts. Establishing milestones and timelines would help ensure CISA does so in a timely manner.
Why GAO Did This Study

Critical infrastructure provides essential functions––such as supplying water, generating energy, and producing food––that underpin American society. Disruption or destruction of the nation's critical infrastructure could have debilitating effects. CISA is the national coordinator for infrastructure protection.

The William M. (Mac) Thornberry National Defense Authorization Act for Fiscal Year 2021 includes a provision for GAO to report on the effectiveness of SRMAs in carrying out responsibilities set forth in the act. This statement addresses (1) how the act changed agencies' responsibilities, and the actions agencies have reported taking to address them; and (2) the extent to which CISA identified and undertook efforts to help agencies implement their responsibilities set forth in the act.

This statement is based on GAO's February 2023 report on SRMA efforts to carry out critical infrastructure protection responsibilities and CISA's efforts to help SRMAs implement those responsibilities. For that report, GAO analyzed the act and relevant policy directives, collected written responses from all 16 sectors using a standardized information collection tool, reviewed other DHS documents, and interviewed CISA officials.

In its February 2023 report, GAO recommended that CISA establish milestones and timelines to complete its efforts to help sector risk management agencies carry out their responsibilities. DHS concurred with the recommendation. Additionally, GAO has made over 80 recommendations which, when fully implemented, could help agencies address their statutory responsibilities.

Germany and Ukraine hit two high-value ransomware targets

The German Regional Police (Landeskriminalamt Nordrhein-Westfalen) and the Ukrainian National Police (Націона́льна полі́ція Украї́ни), with support from Europol, the Dutch Police (Politie) and the United States Federal Bureau of Investigations, targeted suspected core members of the criminal group responsible for carrying out large-scale cyberattacks with the DoppelPaymer ransomware.

This ransomware appeared in 2019, when cybercriminals started using it to launch attacks against organisations and critical infrastructure and industries. Based on the BitPaymer ransomware and part of the Dridex malware family, DoppelPaymer used a unique tool capable of compromising defence mechanisms by terminating the security-related process of the attacked systems. The DoppelPaymer attacks were enabled by the prolific EMOTET malware.

The ransomware was distributed through various channels, including phishing and spam emails with attached documents containing malicious code — either JavaScript or VBScript. The criminal group behind this ransomware relied on a double extortion scheme, using a leak website launched by the criminal actors in early 2020. German authorities are aware of 37 victims of this ransomware group, all of them companies. One of the most serious attacks was perpetrated against the University Hospital in Düsseldorf. In the US, victims payed at least 40 million euros between May 2019 and March 2021.

During the simultaneous actions, German officers raided the house of a German national, who is believed to have played a major role in the DoppelPaymer ransomware group. Investigators are currently analysing the seized equipment to determine the suspect’s exact role in the structure of the ransomware group. At the same time, and despite the current extremely difficult security situation that Ukraine is currently facing due to the invasion by Russia, Ukrainian police officers interrogated a Ukrainian national who is also believed to be a member of the core DoppelPaymer group. The Ukrainian officers searched two locations, one in Kiev and one in Kharkiv. During the searches, they seized electronic equipment, which is currently under forensic examination.
Europol on-site to speed up forensic analysis of seized data

On the action days, Europol deployed three experts to Germany to cross-check operational information against Europol’s databases and to provide further operational analysis, crypto tracing and forensic support. The analysis of this data and other related cases is expected to trigger further investigative activities. Europol also set up a Virtual Command Post to connect the investigators and experts from Europol, Germany, Ukraine, the Netherlands and the United States in real time and to coordinate activities during the house searches. Europol’s Joint Cybercrime Action Taskforce (J-CAT) also supported the operation. This standing operational team consists of cybercrime liaison officers from different countries who work on high-profile cybercrime investigations.

From the beginning of the investigation, Europol facilitated the exchange of information, coordinated the international law enforcement cooperation and supported the operational activities. Europol also provided analytical support by linking available data to various criminal cases within and outside the EU, and supported the investigation with cryptocurrency, malware, decryption and forensic analysis.

NATO and European Union launch task force on resilience of critical infrastructure

Senior officials from NATO and the European Union met to launch a new NATO-EU Task Force on Resilience of Critical Infrastructure. Cooperation to strengthen critical infrastructure has become even more important in light of the sabotage against the Nord Stream pipelines, and Russia’s weaponisation of energy as part of its war of aggression against Ukraine.

First announced by NATO Secretary General Jens Stoltenberg and European Commission President Ursula von der Leyen in January, the initiative brings together officials from both organisations to share best practices, share situational awareness, and develop principles to improve resilience. The Task Force will begin by focusing on four sectors: energy, transport, digital infrastructure, and space.

Announcing the initiative in January, Mr Stoltenberg said: "We want to look together at how to make our critical infrastructure, technology and supply chains more resilient to potential threats, and to take action to mitigate potential vulnerabilities. This will be an important step in making our societies stronger and safer."

NATO-EU cooperation has reached unprecedented levels in recent years, and particularly since the start of Russia’s war of aggression against Ukraine. In January, NATO and EU leaders signed a new joint declaration to take partnership between the organisations to a new level, including on emerging and disruptive technologies, space, and the security impact of climate change.

CISA Launches Ransomware Warning Pilot for Critical Infrastructure

The Cybersecurity and Infrastructure Security Agency publicly announced that it has established a pilot program to identify vulnerabilities within critical infrastructure systems that are known to be exploited by ransomware groups and threat actors.

According to CISA, the ransomware vulnerability warning pilot—or RVWP—will “identify organizations with internet-accessible vulnerabilities commonly associated with known ransomware actors by using existing services, data sources, technologies and authorities, including our free Cyber Hygiene Vulnerability Scanning service.”

The RVWP first began on 30th January when CISA contacted 93 organizations “identified as running instances of Microsoft Exchange Service with a vulnerability called ‘ProxyNotShell,’ which has been widely exploited by ransomware actors.”

“This initial round of notifications demonstrated the effectiveness of this model in enabling timely risk reduction as we further scale the RVWP to additional vulnerabilities and organizations,” CISA said.

The pilot program was created in response to the Cyber Incident Reporting for Critical Infrastructure Act, or CIRCIA, a 2022 law that required CISA “to develop and implement regulations requiring covered entities to report covered cyber incidents and ransomware payments” to the agency. CISA said the RVWP would be “coordinated by and aligned with the Joint Ransomware Task Force,” an interagency body that was also established by CIRCIA.

"Ransomware attacks continue to cause untenable levels of harm to organizations across the country, including target rich, resource poor entities like many school districts and hospitals,” Eric Goldstein, CISA’s executive assistant director for cybersecurity, said in a statement. “The RVWP will allow CISA to provide timely and actionable information that will directly reduce the prevalence of damaging ransomware incidents affecting American organizations.”

Ransomware Attacks on Critical Infrastructure Fund DPRK Espionage Activities

CISA, the National Security Agency (NSA), the Federal Bureau of Investigation (FBI), the Department of Health and Human Services (HHS), and Republic of Korea’s Defense Security Agency and National Intelligence Service have released a joint Cybersecurity Advisory (CSA), Ransomware Attacks on Critical Infrastructure Fund DPRK Espionage Activities, to provide information on ransomware activity used by North Korean state-sponsored cyber to target various critical infrastructure sectors, especially Healthcare and Public Health (HPH) Sector organizations.

The authoring agencies urge network defenders to examine their current cybersecurity posture and apply the recommended mitigations in this joint CSA, which include:

- Train users to recognize and report phishing attempts.
- Enable and enforce phishing-resistant multifactor authentication.
- Install and regularly update antivirus and antimalware software on all hosts.

See Ransomware Attacks on Critical Infrastructure Fund DPRK Espionage Activities for ransomware actor’s tactics, techniques, and procedures, indicators of compromise, and recommended mitigations. Additionally, review StopRansomware.gov for more guidance on ransomware protection, detection, and response.

1 2 3 4 5 30