Mythical Beasts and Where to Find Them: Mapping the Global Spyware Market and its Threats to National Security and Human Rights

Executive Summary

Despite its contribution to human rights harms and national security risks, the proliferation of spyware remains rife. A significant channel for this proliferation is sale through a global market, of which most public information is known about only a handful of vendors. While some of these entities have achieved infamy, like NSO Group and the Intellexa Consortium, most others have largely flown under the radar.

The Mythical Beasts project addresses this meaningful gap in contemporary public analysis on spyware proliferation, pulling back the curtain on the connections between 435 entities across forty-two countries in the global spyware market. These vendors exist in a web of relationships with investors, holding companies, partners, and individuals often domiciled in different jurisdictions.

This market is a significant vector for facilitating the human rights harms and national security risks posed more broadly by spyware, software that facilitates unauthorized remote access to internet-enabled target devices for purposes of surveillance or data extraction. It is possible for policymakers to make significant progress in limiting these harms and risks by influencing this market, rather than playing “whack-a-mole” with individual vendors or transactions. This progress is possible now, even in the face of basic disagreements over what constitutes a “legitimate” use of spyware. Besides changes to participants in the market, greater transparency will also support more effective policies related to spyware, rooted in cooperative international action.

Developed as part of a wider study of proliferation and international cybersecurity, this report provides an analysis of the accompanying dataset comprised of information from 1992 to 2023 on forty-nine vendors along with thirty-six subsidiaries, twenty-four partner firms, twenty suppliers, and a mix of thirty-two holding companies, ninety-five investors, and one hundred and seventy-nine individuals, including many named investors. There are six trends that hold for this dataset, a detailed but even still incomplete sample: 1) concentration of entities in three major jurisdictions (Israel, Italy, and India), 2) serial entrepreneurship across multiple vendors, 3) partnerships between spyware and hardware surveillance vendors, 4) regularly shifting vendor identities, 5) strategic jurisdiction hopping, and 6) cross-border capital flows fueling this market.

These trends inform a set of policy recommendations to produce greater transparency across the market, limit the jurisdictional arbitrage of vendors seeking to evade limits on their behavior, and more effectively scrutinize supplier and investor relationships.

**Figure 1: Policy recommendations to produce greater transparency across the market**

Commercial acquisition of spyware is not the root cause of its abuse. While this project is focused on bringing transparency to participants in this market, it does not argue that only transactions through this market pose proliferation harms or risks. An information gap exists in what is known about the spyware market and its varied participants, a gap that is impeding international cooperation on policies that could meaningfully reduce the harms and risks posed by spyware. This report seeks to offer new data and analysis to bridge that gap and support the work of researchers and policymakers more widely.

Introduction

For at least the last thirty years, “mythical beasts” have been lurking around the globe, assuming the names of varying species of fish, fowl, and other creatures rooted in lore. These mythical beasts—often with dramatic naming conventions—are spyware: software that facilitates unauthorized remote access to an internet-enabled target device for purposes of surveillance or data extraction. The companies that sell these tools are sustained by an increasingly diverse array of government customers across a global market, even in the face of scattered regulatory efforts targeting spyware supply chains.¹

Out of 195 countries in the world, at least eighty are known to have procured spyware from commercial vendors.²

Fourteen of the twenty-seven countries in the European Union have purchased spyware from just one vendor, the NSO Group.³

Spyware vendors were attributed to fifty percent of all zero-day exploits discovered by one company’s threat research team in 2023, including sixty-four percent of all exploits in mobile and browser software.⁴

While the annual revenue generated by this market is unknown and subject to repeated speculation, largely recycling the same unsourced statistic, at least one vendor has considered an initial public offering valuation of $2 billion.⁵

With the proliferation of spyware, from NSO Group’s Pegasus to Intellexa Consortium’s Predator, comes increased attention to its use. Some argue that spyware can be employed as a legitimate law enforcement and intelligence tool.⁶ It has also been used by states to extend surveillance power well beyond their physical borders, making it easier to track, arrest, kidnap, and even kill their citizens.⁷ In these abuses of spyware, the victims are most often journalists, activists, opposition politicians, and a myriad of other individuals whose activity has attracted hostile interest from their governments. For years, civil society organizations like AccessNow and Amnesty International have sought to bring attention to these abuses and have reported on spyware’s use on nearly every continent.⁸

State surveillance, harassment, repression, and outright murder predate spyware, and there is little to suggest spyware “causes” these abuses. Measuring the human rights harms and national security risks of spyware against its value to law enforcement or intelligence activities is also challenging as these activities are, by their nature, even less visible. Few governments have sought to demonstrate the range of legitimate uses of spyware or its impacts. As a result, when considering spyware’s effects on society, there is a bias to what is known.

Still, what is known is an abundance of public evidence of the totality of abuses made easier— perhaps even directly possible—by spyware.⁹

It is not in dispute that spyware makes it easier for states to penetrate even the most robust commercial technologies, cell phones, computers, and communications services; makes it far easier to act against citizens beyond state borders; and even provides governments with the ability to target senior officials, both domestically and abroad, where they might otherwise have no means to do so.¹⁰ Where that information is used to facilitate repression and abuse, its harms are untenable. Where that information is gathered and used subject to due diligence and effective oversight in pursuit of credible law enforcement and intelligence activities subject to the limits of the law, its effects may provide for the public interest. These two categories overlap and are altogether too often separated by good intent and cursory legal review.

The proliferation of spyware also poses national security risks as it makes it more likely for states to become “more capable—for instance while conducting cyber-espionage for commercial or intelligence gain—or ready for more disruptive or damaging operations.”¹¹The proliferation of these capabilities in most states takes place with few effective restraints, strict controls, or meaningful oversight mechanisms. This is a recognized policy challenge and one which has been taken up in various forms by some governments, largely in Europe, the US, and the UK.

Briefly Through the Past

Digital surveillance technologies, which include spyware, are known as dual-use goods, meaning they can be “used for both defense and civilian purposes.”¹² Dual-use technology in forty-two countries falls under a multilateral export control regime established in 1996, the Wassenaar Arrangement on Export Controls for Conventional Arms and Dual-Use Goods and Technologies (the Wassenaar Arrangement).¹³ In 2013, the Wassenaar Arrangement was amended to include “intrusion software” but after considerable feedback from the security research community and significant delay, the language was revised.¹⁴ While the Wassenaar Arrangement is not legally binding, its signatories typically voluntarily implement its control list into domestic regulations, oftentimes requiring firms whose products are listed to acquire special licenses to export these items.

Within the European Union, export controls are governed by the EU’s Dual-Use Regulation.¹⁵ The first EU legislation on dual-use goods was enacted in 1994, underwent significant changes in 2009, and a new version was enacted in 2021 to implement and modernize the EU’s export control regime.¹⁶Member states are required to abide by this common set of restrictions but may introduce additional controls on non-listed dual-use items due to public security or human rights considerations.

The United States is also a participant in the Wassenaar Arrangement and the Bureau of Industry and Security (BIS) within the Department of Commerce has the authority to regulate dual-use exports by issuing export licenses.¹⁷The BIS Entity List contains names and organizations subject to specific additional license requirements.¹⁸

In 2021, BIS added four entities to this list, including for the first time two spyware vendors:

Candiru Ltd

NSO Group

And two suppliers:

COSEINC

Positive Technologies AO.¹⁹

In 2023, BIS also added four companies associated with the Intellexa Consortium to the Entity List: Intellexa S.A., Cytrox AD Holdings ZRT, Intellexa Limited (Ireland), and Cytrox AD (North Macedonia) as they were determined by BIS to be “trafficking in cyber exploits used to gain access to information systems, threatening the privacy and security of individuals and organizations worldwide.”²⁰

In addition to export controls, the European Union and the United States have sought to implement other measures to limit spyware proliferation. In 2022, in response to the investigative findings of the Pegasus Project, an international investigative journalism initiative, the European Parliament established the PEGA Committee to investigate the misuse of surveillance spyware, including the NSO Group’s Pegasus and similar spyware services.²¹ The committee concluded that European Union governments abused spyware services, lacked necessary safeguards to prevent misuse, and in at least one jurisdiction, Greece, the government had facilitated the export of Predator spyware which was then itself abused, here by Sudan’s Rapid Support Forces militias, who are reported to have committed war crimes.²² Despite the committee’s recommendations, the EU has not adopted any legislation as a bloc to curb the development or sale of spyware.²³

More Recently

The last twenty-one months saw a surge in policymaking activity building on these recent efforts. Most visible has been from the United States, which has enacted punitive measures targeting entities selling spyware and driven some measure of diplomatic consensus to “recognize the threat posed by the misuse of commercial spyware” and acknowledge the “fundamental national security and foreign policy interest in countering and preventing the proliferation of commercial spyware.”²⁴

In March 2023, the United States first proposed blocking US government agencies from using “commercial spyware.” Under Executive Order 14093, the Biden administration prohibited the operational use of commercial spyware that presents a significant threat to national security.²⁵ Also in March of 2023, the US and several other countries signed The Joint Statement on Efforts to Counter the Proliferation and Misuse of Commercial Spyware, pledging to “work collectively to counter the proliferation and misuse of commercial spyware.”²⁶

In March 2024, the US Department of Treasury Office of Foreign Assets Control levied sanctions against several entities, some of which are also listed on the BIS Entity List.²⁷ Ultimately Treasury sanctioned:

Tal Dilian

Sara Hamou

Intellexa S.A.

Intellexa Limited

Cytrox AD

Cytrox Holdings Crt

Thalestris Limited²⁸

So far, the US has refrained from sanctioning five other entities within the Intellexa Group, previously identified publicly, and perhaps others, including entities associated with Thalestris Limited.²⁹ That same month, several additional countries joined as signatories in an expansion of The Joint Statement on Efforts to Counter the Proliferation and Misuse of Commercial Spyware.³⁰

In April 2024, the US Department of State announced a visa restriction policy to “promote accountability for the misuse of commercial spyware.”³¹This extended statutory language from 2021, originally implemented as visa bans on “individuals who, acting on behalf of a foreign government, are believed to have been directly engaged in serious, extraterritorial counter-dissident activities, including those that suppress, harass, surveil, threaten, or harm journalists, activists, or other persons perceived to be dissidents for their work, or who engage in such activities with respect to the families or other close associates of such persons.”³²The new restrictions pertain to individuals who have been involved in the development and sale of commercial spyware and their immediate family members.³³ Thirteen individuals, whose identities are not public, have been subject to this action as of the date of this writing.³⁴

A new multilateral effort from the UK and French governments, the Pall Mall process, has also brought together an even wider array of state and non-state participants to develop principles, and perhaps practical policy action, to counter spyware proliferation.³⁵ Pall Mall includes non-state groups and a much wider set of states than the Joint Statement signatories but, so far, with much more ambiguous outcomes, including a debated set of principles and plans for a broad consultative process.

Collectively these US and allied efforts demonstrate there is a growing focus on curtailing the proliferation of spyware. However, still missing from these discussions is a common picture of the spyware market with sufficient detail to understand the diversity of market participants and relationships that stretch across borders.

A Turn to the Spyware Market

This report offers a new dataset covering 435 entities (incl. forty-nine vendors and twenty suppliers) across the spyware market. The data spans forty-two different countries and nearly thirty years, covering vendors, investors, and other corporate relationships. The collection of this data and its publication is an attempt to address a systematic bias toward the operations of a small handful of well-known firms that inform assumptions about the interactions and relationships of a large global market.

This narrow focus has helped obscure the impact of dozens of other vendors and the importance of their relationships with both investors and suppliers of crucial software components, including working exploits of some of the most widely used software (e.g. iOS, Android). This project supports the “turn to spyware” in recent transatlantic policy activity and should enable a more robust, market-first (rather than vendor-centric) approach. Such an approach can leverage both conventional tools to constrain and shape markets as well as new policies to address the unique dimensions of spyware.

Unlike other more tightly regulated markets, the market for spyware lacks public data that is consistent, reliable, and clearly sourced. While a single export authority, such as Israel’s Defense Exports Control Agency, may track sales out of the country, this information is neither public nor combined with similar resources from other jurisdictions. Thus, the view of the spyware market is limited even for exceptionally well-equipped states. Researchers, journalists, and policymakers alike must scrape through a variety of different resources just to scratch the surface of this market that has cloaked itself in secrecy, making it difficult for policy action. There is some comparison even to be had with the approach of US policymakers in regulating the market for cannabis over the last several decades. Once a widely banned yet still pervasively acquired substance, the cannabis market has now been legalized in many states and is subject to enormous scrutiny. This new approach focused not on blocking cannabis transactions or consumption but rather on leveraging market forces, accepting some legitimate use, and creating parameters for responsible procurement. Proliferation will not be prevented by a well-regulated and more transparent market, but it can be better channeled and made subject to controls, less opaque, and less harmful.

Policy responses to address spyware as a market is preferable over advocacy for a complete ban. Such a ban would likely supercharge government calls for exceptional access to encrypted services and data while sapping momentum toward better approaches to spyware in the US and UK, home to two of the most active policymaking communities on this issue.

An important caveat: the dataset collected here does not catalog use, so the authors cannot make novel claims about what constitutes “legitimate use” of spyware. Few of the policies mentioned in the previous section define legitimate use with sufficient precision to enforce granular bans on behavior and even these are not transparent.³⁶ The lack of a common understanding of the scale, diversity, and relationships within the spyware market is a barrier to effective policymaking. Thus, this report argues for how to improve transparency in the market and develop sufficient granular controls to enable these kinds of distinctions in use.

The next section of this report explains the methodology associated with this dataset as well as definitions and a summary of the data. The following section analyzes major trends from this data and highlights specific examples and implications for policy before developing specific recommendations. As the authors have written previously, “markets matter,” and this report argues that the current surge in policymaking toward spyware can best be sustained and made more impactful by focusing policymaking on the spyware market instead of just a handful of the most well-known firms.³⁷

Methodology, Definitions, and Navigating the Dataset

This section discusses the project’s scope, including a working definition of spyware, data collection methodology, and sources, and closes with the major definitions and terms used as part of this dataset’s coding framework.

Spyware: is a type of malicious software that facilitates unauthorized remote access to an internet-enabled target device for purposes of surveillance or data extraction.³⁸ Spyware is sometimes referred to as “commercial intrusion [or] surveillance software” with effectively the same meaning.³⁹ This research considers the “tools, vulnerabilities, and skills, including technical, organizational, and individual capacities” as part of the supply chain for spyware and the meaningful risks posed by the proliferation of many of these components.⁴⁰

This project is concerned with the commercial market for spyware and provides data on market participants. Focusing on the market does not presume that all harms from spyware stem from how it is acquired, or whether that acquisition is a commercial transaction with a third party (versus developed “in-house” by the customer). Some definitions of spyware differentiate it by the means with which it is acquired, creating confusion over the fundamental distinction between “spyware” and, for instance, “commercial spyware.”⁴¹

… so “Commercial” Spyware?

Transactions across the spyware market may be less regulated than in-house development of spyware but they are far from the only source of harm and insecurity. Policies that seek only to mitigate harms from the commercial sale of these capabilities risk ignoring their wider harms and avoid the opportunity to address fundamental concerns over surveillance and the full spectrum of government uses of these technologies.

The debate over what constitutes legitimate uses of spyware is ongoing, but commercial sale is a poor proxy for the degree of responsible or mature use. History has shown that this market is only one, albeit significant, part of a wider proliferation challenge.⁴²Many human rights violations associated with spyware occur in the context of their use for state security purposes (e.g., by intelligence agencies), highlighting the diverse harms and risks posed by the proliferation of spyware. These include what some researchers have termed “vertical” uses (by states against their own populations) and “diagonal” uses (against the population of other states, including diaspora).⁴³ There is some normative loading in the term “spyware” vs. the more functional “malware” or the rather impenetrable “commercial intrusion capabilities” but is beneficial to have a common term of art in many of these debates.

This report and accompanying dataset are mainly inclusive of investigations into vendors and suppliers that have been found selling spyware to governments across the world that have then used this software to abuse human rights. However, this is only one side of the coin. Far less data exists on the use of spyware for a myriad of intelligence and counterintelligence purposes, including “national security” missions both genuine and troubling. The report cannot resolve these tensions but does seek to frame them in service of a more immediate and practical purpose—and a better understanding of the market that provides the software tools and services to carry out these acts.

Commercial acquisition of spyware is not the root cause of its abuse. While this project is focused on bringing transparency to participants in the spyware market, it does not argue that only transactions through this market pose proliferation risks or harms.⁴⁴To avoid further confusion in both analysis and policy the authors do not include the term “commercial” in the definition of spyware. While the debate continues about how to manage these risks, this project sheds better light on those buying, selling, and supporting this market.

A Final Note on Scope

Spyware works without the consent or knowledge of the target or others with access to the target’s device; thus, this report does not consider the market for so-called “stalkerware,” which generally requires physical interaction from an individual, most often a spouse or partner, with access to a user’s device.⁴⁵ This definition also excludes software that never gains access to a target device, such as surveillance technologies that collect information on data moving between devices over wired (i.e., packet inspection or “sniffing”) or wireless connections. This definition also excludes hardware such as mobile intercept devices, known as IMSI catchers, and any product requiring close or physical access to a target device, such as forensic tools.⁴⁶

This definition is limited, by design, to disentangle lumping various other surveillance toolsets into the definition of spyware.

Building the Dataset

This dataset represents a meaningful sample of the market for spyware vendors, but it is not a complete record and this report can only speak to trends and patterns within this data, not the market as a whole. The data is confined to entities for which there is a public record (i.e. registered businesses) and for which public information links the vendor to the development or sale of spyware or its components.⁴⁷

To develop a list of vendors, the authors started by creating an initial “most visible” list of those with the widest public exposure from the use of their wares, relying principally on public reporting from Amnesty International, Citizen Lab, and the Carnegie Endowment for International Peace, as well as public reporting from a variety of news outlets. This initial set of vendors was the starting point for searching public corporate registries and a mix of public and private-sector corporate databases to profile each company in greater depth and find additional connections.

All the vendors identified through this process were included if they 1) publicly advertised products or services that matched the above definition of spyware, 2) were described as selling the same products by public reporting in the media or by civil society researchers, or 3) showed evidence of the products through court records, leaks, or similar internal documentation. As part of this search process, the team gathered records on subsidiaries and branches associated with each vendor, their publicly disclosed investors, and, where possible, named suppliers.

Each entity identified in this process was identified by at least two different open sources. In all cases for which data is available, the dataset includes vendor activities from the start of operation until 2023, or until records indicate that the vendor’s registration had ceased in a jurisdiction. The sources of public information on both firms’ activities and their organization varied but largely stemmed from different forms of corporate registration, records, and transaction data.

Defining Entities in the Spyware Market

The summary profiles of vendors and suppliers can be found in the Trends section of the report, as well as in Appendix A.

Figure 2: Entities in the dataset are spread across jurisdictions

This section identifies and explores six distinct trends from the data gathered in this sample of the spyware market:

Analysis

The majority of identified entities in this sample are domiciled in Israel, India, and Italy
Serial entrepreneurs are rife
The robust partnerships between spyware and hardware surveillance vendors
The deliberate and repeated efforts by firms to shift their identities and even corporate structure
The movement of those corporate structures across strategic jurisdictional boundaries
The significant cross-border mobility of capital supporting spyware development and sales

The dataset is based on a collection of known or reported spyware vendors and analysis, using these as a basis to identify first-order connections and map the resulting network of entities. We included only those entities as defined above. As a result, the dataset is a sample of the spyware market and the trends speak to this data.

1. The Three I’s

Across the 435 entities in this sample of the spyware market, there is a significant concentration of vendors and associated entities in three jurisdictions: Israel, India, and Italy. These states are by no means the only hosts of spyware vendors or their investors and partners, but they are unusually prolific.

Israeli cluster: Eight vendors (NSO Group, SaitoTech—formerly Candiru Ltd, Cognyte, Paragon Solutions, MerlinX, Quadream Inc./InReach Technologies Limited, Blue Ocean Technologies, and Interionet.) This cluster comprises 43.9 percent of the entities in this dataset. The average period of activity (from the time of initial registration to the most recent) for each of these vendors is 6.6 years.

Indian cluster: Five vendors (Aglaya Scientific Aerospace Technology Systems Private Limited, Appin Security Group, BellTroX Infotech Services Private Ltd., CyberRoot Risk Advisory Private Limited, and Leo Impact Security Service PVT Ltd.) as well as one supplier (RebSec Solutions). This cluster covers 7.8 percent of the entities in this dataset, with the average period of activity for vendors lasting 10.1 years.

Italian cluster: Six vendors (Dataflow Security s.r.l., DataForense s.r.l., Memento Labs srl—formerly Hacking Team srl or Grey Heron, Movia SPA, Negg Group/Negg International, s.r.l., and RCS ETM Sicurezza S.p.A.) and one supplier (VasTech). This cluster includes 13.6 percent of all entities in the dataset with an average period of activity lasting 6.1 years.

This dataset represents a global market and it is notable that, for all the press on some firms in the country, Israel represents barely half of this sample. It is also important to note that in the early stages of this project, the most widely reported vendors were based in Israel, which means they constituted a much larger portion of early versions of the dataset. Italy is a notable jurisdiction given its home in the EU, where debates continue about how to govern the presence and operation of spyware vendors. The geographic spread of these “Three I’s” underscores the need for cooperative approaches to driving transparency and shifting behavior in the spyware markets and highlights the absence of both Israel and India in the most recent high-profile spyware policymaking process, the Pall Mall declaration.⁴⁸

2. Serial Entrepreneurs

Across this sample of the market, there is a recurring pattern of employees, including founders leaving their first firm to found and work in other companies, often repeatedly. This is not unlike other startup cultures where this kind of serial entrepreneurship is common. It is interesting in the context of this market, however, given the essential similarity of these products’ intended function and the assumed stickiness of customer relationships with founders and senior employees. Within the dataset, founders of vendors and suppliers are involved in 2.2 companies on average.

**Figure 3: Employees frequently make the jump from employee to founde**r

The NSO Group, arguably the most well-known spyware vendor, is a prime example of this phenomenon. The firm was founded in 2010 in Israel by Niv Karmi, Omri Lavie, Shalev Hulio, and Eddy Shalev and is the developer of the Pegasus spyware. Despite investigations from the EU and regulatory action from the US, the firm continues to operate branches in the United States and Luxembourg along with subsidiaries in Bulgaria and Cyprus.⁴⁹

The vendor Quadream Inc., known for the spyware Reign, was founded in 2016 in Israel by former NSO Group employees Guy Geva and Nimrod Reznik, as well as former military official Ilan Dabelstein and goes by the name Kvader Ltd. in Israel.⁵⁰ Like Quadream Inc., Interionet Systems Ltd. (Interionet) is an Israeli vendor founded in 2015 by Yair Pecht and Sharon Oknin, former employees of NSO Group.⁵¹ Israeli businessman Joshua Lesher, an NSO shareholder and board member, also sits on Interionet’s board.⁵²

Interionet “develops malware for internet routers” and is notable for compromising internet-of-things devices, such as video surveillance cameras.⁵³ In 2022, Interionet won a contract with the Belgian police for their €299 million modernization project called I-Police.⁵⁴ In a study of cyber capabilities in the international arms market⁵⁵the authors assessed with high confidence that Interionet “is willing to market its capabilities in countries which are not allied to the American and European interests.”⁵⁶

**Figure 4: Mapping connections in the NSO Group, Quadream, and Interionet clusters**

There are examples of this trend outside of Israel as well. The Appin Security Group, established by Rajat Khare and his brother Anuj, has since been alleged to have targeted and spied on entities worldwide. Materials online appear to show the company offering hack for hire services.⁵⁷ BellTroX Infotech Services Private Ltd. was registered in India in 2013 by Sumit Gupta, formerly an employee of Appin Security Group.⁵⁸ BellTrox Infotech Services Private Ltd. has been previously named by Meta as offering “hack for hire” services.⁵⁹

**Figure 5: Mapping connections in the Appin Security Group and BellTrox Infotech Services Private Ltd Clusters**

This suggests that more closely governing the talent pools of individuals pivoting between companies might restrict these individuals from creating their own companies and limit the proliferation of spyware vendors. To impede talent from pivoting easily between vendors, export licensing bodies could require more detailed information on key personnel and their past employment to help identify serial offenders of the laws and policies of other jurisdictions. Policymakers should also consider focusing on individuals when attempting to limit harmful activities by a vendor, rather than just the vendor as a business entity, given the fluidity of talent between firms.

3. Partnership with Hardware Surveillance

Spyware vendors in the dataset have sometimes partnered with hardware-based surveillance companies whose products might complement the functionality of their spyware tools. We have identified nine vendors or suppliers known to have at least one partner, with at least five vendors partnering with at least one hardware company. The most active example of these partnerships is the Intellexa Consortium, encompassing relationships between seven distinct hardware firms.⁶⁰Formed around founder Tal Dilian in 2018, the Intellexa Group comprises several companies including Cytrox AD, WS WiSpear Systems Limited (also founded by Dilian), and Senpai Technologies Ltd.⁶¹ In 2020, Intellexa Group expanded to include Intellexa S.A., formerly known as Intellexa Single Member.⁶² Cytrox AD was formed in 2017 by Rotem Farkash and Abraham Rubinstein in North Macedonia and developed the spyware known as Predator. WS WiSpear Systems Limited specializes in intercepting targeted Wi-Fi signals and extracting passwords and communications at long range, and Senpai Technologies Ltd. is an open-source intelligence company that specializes in analyzing data from phones infected with spyware.⁶³

In addition to the Intellexa Group, there is the Intellexa Alliance, formed in 2019 as a partnership between the Intellexa Group and the Nexa Group. Nexa Group is a cluster of four other companies selling interception technology that retail their products together.⁶⁴ It remains unclear whether the Intellexa Alliance is still operational, as tensions have emerged between the two entities.⁶⁵ Together, the Intellexa Group and Intellexa Alliance comprise the Intellexa Consortium, profiled in more detail in the Cyber Statecraft Initiative’s earlier report, “Markets Matter: A Glance into the Spyware Industry.”⁶⁶

**Figure 6: Mapping connections in the Intellexa Consortium and Nexa Group clusters**

This trend appears in the Italian market as well, with vendor Memento Labs srl (subsequently known as Hacking Team srl) partnering with South African firm VASTech, founded in 1999 by Frans Dreyer, to develop a passive interception product for wireless communications (building on earlier work from the firm DataVoice).⁶⁷ VASTech maintains two offices in South Africa while VASTech AG operates in Switzerland, and VAS Technologies is located in the UAE.⁶⁸ VASTech would later go on to propose a partnership with Hacking Team srl to directly resell the vendor’s spyware in 2015.⁶⁹

**Figure 7: Mapping connections between Hacking Team srl and VASTech clusters**

Firms fostering relationships with others offering complementary products is not novel but it is nonetheless interesting to see in this sample of the spyware market. The phenomenon underlines the further importance of policies that address the market as a whole and collaboration across multiple states, as vendor or jurisdiction-specific actions often have limited effect on these wider relationships. Regulating the kinds of support provided to spyware vendors selling to government agencies could help govern the kinds, and content of these partnerships and extend important transparency measures like “Know Your Vendor” requirements to important firms a step beyond the initial spyware transaction. This recommendation is particularly important due to shifting vendor identities.

This trend also highlights the potentially complex relationship between the spyware market and vendors of other electronic surveillance technologies. An open question for further research is how efforts to constrain spyware sales may impact these complementary tools.⁷⁰ A further question raised is how substitutable these non-spyware alternatives might be for existing customers and the extent to which spyware firms (like VASTech) offer both spyware and other products to diversify and strengthen their business.

4. Shifting Vendor Identities

Spyware vendors will change legal names and even shift entire corporate structures, which can serve to obscure their identity and, potentially, manage the impact of negative reporting.

Despite name changes, reporting often refers to entities by their most popularized name. This can obscure the vendor’s ongoing activity and impede researchers, policymakers, and any firms attempting to exercise due diligence in potential investments. On average, the entities tracked in the dataset changed names more than once, an average of 1.4 times over the time observed, with a name lasting an average of 4.5 years vs. an average vendor lifespan of nearly double that length.

To put this into perspective across the rest of the dataset, 14.3 percent of the vendors underwent a name change while 10.2 percent of all entities changed their name (excluding individuals). Holding Companies had the highest percentage of name changes at 34.4 percent, followed by Vendors (14.3 percent), Partners (20.8 percent), Suppliers (five percent), and Investors (2.1 percent).⁷¹

Figure 8: Entities change their legal name to obscure their identity and manage the impact of negative press

The most persistently shifting identity is that of the firm originally known as Candiru Ltd, which changed its name four times over the ensuing nine years, and is known at the time of this writing as Saito Tech Ltd.⁷² The vendor originally known as Candiru Ltd was incorporated in 2014 in Israel by founders Ya’acov Weitzman and Eran Shorer.⁷³ Candiru Ltd has sold products to Hungary, Spain, and the United Arab Emirates, who all used the spyware for political suppression of opposition and civil society.⁷⁴ The group’s annual name changes between 2016 and 2020 did not come with changes to the corporate structure. In 2021, Candiru Ltd and its associated names were added to the US Entity List alongside NSO Group.⁷⁵ In popular discourse, the vendor is often called Candiru Ltd, this report refers to all vendors by their present legal name, which for Candiru Ltd is Saito Tech Ltd.

On the other hand, Memento Labs srl, initially named Hacking Team srl, retained its original brand for sixteen years, the longest of any entity in the dataset, until changing it in 2019. Formed in 2003 in Italy by David Vincenzetti and Valeriano Bedeschi, Hacking Team srl developed the Remote Control Systems (RCS) spyware. A wide breadth of information is available on the business model of Hacking Team srl due to a leak of its internal data in 2015.⁷⁶Hacking Team srl has been reported to sell to Ecuador, Nigeria, and Saudi Arabia, as well as many states, all of which may have utilized the RCS spyware to suppress human rights.⁷⁷ Despite legal obstacles, including the revocation of the firm’s export license in 2016, Hacking Team srl continued to exist as a company. From 2017 to 2018, there was also a potential spin-off of Hacking Team srl known as Grey Heron, as announced by a Hacking Team srl representative at a security conference in the United Kingdom.⁷⁸The company was officially renamed Memento Labs srl in 2019 after being acquired by InTheCyber Group fSA, a Switzerland-based investor, in an effort to rebrand itself.⁷⁹

Another example of this behavior can be observed in the Indian vendor Appin Security Group. Beginning in 2014, Appin Technology Ltd., Appin Security Group’s parent company, began a rapid succession of name changes evolving from Appin Technology Ltd. to Mobile Online Order Management Private Limited, then from this name to Chemieast Engineering, and then from Chemieast Engineering to Sunkissed Organic Farms. Appin Security Group itself also changed names to Approachinfinate Computer and Security Consultancy Grp and then to Adaptive Control Security Global Corporate.“⁸⁰ This approach echoes Saito Tech’s approach of rapid name changes without significant alterations to business structures. Equus Technologies provides a good example of name changes in response to press and reporting. Founded in 2014 by Matan Markovics, Daniel Hanga, and Tal Tchwella in Israel, Google attributed the firm as the developer of the Lipizzan software in 2017 and labeled the vendor a “cyber arms company.”⁸¹ After this reporting, Equus Technologies struggled to recover from reputational damage as it started losing customers and shareholders shrank their positions in the company.⁸² Equus then changed its name to MerlinX between 2017 and 2018.⁸³ Tal Tchwella, one of its three founders, also left the company.⁸⁴ MerlinX was later acquired by Bindecy, an Israeli company specializing in vulnerability research, in 2021.⁸⁵

Figure 9: Charting entity name changes over time

These various examples show why it is difficult for policymakers and researchers alike to keep track of vendors, creating an illusion that a vendor has ceased operations when they are functioning under a different name. This image grows more complicated with subsidiaries and branches as they too may shift names rapidly, furthering an already opaque market.

To counter this trend, policy solutions can emphasize individual and investor relationships. A baseline improvement for spyware procurement would also be mandatory “Know Your Vendor” requirements to disclose first- and second-order supplier relationships. Better and more consistent transparency from corporate registries would also help establish the link between these identities, even across jurisdictions—which the report turns to next.

5. Strategic Jurisdiction Hopping

Several of the vendors captured in the dataset appear to have constructed subsidiary, branch, and partner relationships that cross strategic jurisdictional boundaries. These relocations may offer a variety of location-specific benefits, from facilitating sales to the EU market with an EU-domiciled firm to situating branches in states with more forgiving laws.

Figure 10: Number of Jurisdictions Per Cluster

In 2017, the Israeli vendor Quadream Inc. set up a supplier, InReach Technologies Limited, in Cyprus which Quadream Inc. claimed in a later court filing was for the “sole purpose of promoting Quadream Inc. products within the European Union.”⁸⁶ InReach Technologies Limited’s financial structure also included A.I.L Nominal Services Ltd. (A.I.L), similarly established in Cyprus in 2010 as a holding company, with an individual with a relationship to the Ministry of Defense.

Quadream Inc. and InReach Technologies Limited’s relationship deteriorated in 2020 and they became entangled in a court case against one another.⁸⁷ While the relationship was strained, it is unclear whether the companies formally separated by the time Citizen Lab released Quadream Inc.’s toolkit in 2023, exposing the company’s capabilities. This led to the company reporting that it would be shutting down operations although the company is still registered in Israel.⁸⁸

The Intellexa Consortium provides another example of this jurisdictional hopping. One investigation discovered through leaked documents shows how the organizer of the Intellexa Consortium, Tal Dilian, and his partner Sara Hamou, utilized Cyprus as a hub for the Predator spyware to gain access to the European market.⁸⁹

Memento Labs srl (formally known as Hacking Team srl) provides an interesting exception to this trend as its founders appear to have worked to make it strictly an Italian-based vendor. Like models found in other businesses that boast national pride, Hacking Team srl is proud to be “Made in Italy.”⁹⁰Their investor base is also mainly Italian, with only two other European countries (Cyprus and Switzerland) present.⁹¹

Figure 11: Vendors construct subsidiary, branch, and partner relationships that hop across strategic jurisdictional boundaries

National laws governing the behavior of subject firms are largely premised on the common recognition by both the regulators and the regulated of sovereign boundaries. These boundaries delimit the application of law between, say, France and the United Kingdom. The deliberate construction of branch and subsidiary relationships to cross these boundaries may offer firms a measure of protection from regulatory approaches like export controls and create significant opacity in their operations and supply chains by wrapping even loose cross-border relationships in the cloak of “internal” corporate activity. The pivot of vendors to less restrictive jurisdictions reduces the efficacy of export controls and policy action must better limit the effects of jurisdictional arbitrage. But this trend is not limited to corporate organization, indeed it is reflected in capital flows as well, leading to the final trend.

6. Money From Across the World Fuels the Spyware Market

Investors in spyware vendors often cross borders with their capital. Like more conventional markets, spyware vendors and suppliers feature an investor base domiciled in many different jurisdictions. Investment in spyware vendors and suppliers is understudied, despite being a factor in the proliferation of this technology. The sample of the spyware market captured by this dataset includes ninety-five investors identified to date. Among these, outside of investors for whom location was not listed, four jurisdictions were most frequently represented: Italy, Israel, the United States, and the United Kingdom—comprising 46.3 percent of all investors.

The character of spyware investment varies from venture capital, private equity, and government loans to outright acquisitions and direct equity ownership. On average in this dataset, each vendor and supplier had 4.75 identified investors, with Figure 12 highlighting where investors are domiciled and to which cluster they invest. For example, the dataset documents fourteen different US entities investing in spyware vendors or suppliers, the bulk of whom (in twelve of the fourteen cases) are based in Israel. Of note, the Israeli and Italian investors captured in this dataset were likely to mostly invest in their own markets versus the United States and the United Kingdom, whose investors largely sent their capital abroad.

Figure 12: Money flows from investor to vendors, often crossing borders

A specific example of this trend is Paragon Solutions. Paragon was established in 2019 in Israel by Ehud Schneorson, Idan Nurick, Igor Bogudlov, and Liad Avraham.⁹²A few years later in 2022, the firm established Paragon Solutions US, a US-domiciled subsidiary.⁹³ Since its establishment, Paragon has made deliberate efforts to break into the US market. Paragon Solutions also has two US-based investors. Battery Ventures, considered to be one of the world’s top venture capital firms and headquartered in Boston, is an investor in Paragon Solutions as of this writing.⁹⁴The listing does not name “Paragon Solutions US”. It is also supported by Blumberg Capital, another large US venture capital firm.⁹⁵

Saito Tech Ltd and NSO Group also have investors domiciled in foreign jurisdictions. Saito Tech Ltd had the US-based Founders Group and NSO Group’s current and past investors include those domiciled in the United Kingdom, like Novalpina Capital, and in the United States, like Francisco Partners Management, Berkely Research Group, and Blackstone Group LP.⁹⁶The transition in NSO Group investors from Novalpina Capital to other investors took a few years and is well covered in the press.

These cross-border capital flows highlight the importance of international cooperation, and perhaps the central role of the United States and EU, in applying more granular controls and scrutiny on investor relationships in the spyware market.

Improving corporate transparency requirements, such as the US’s recent move to compel companies to report their beneficial owners in line with policies in other countries, will support improved investor due diligence and deal review inside the United States.⁹⁷ For vendors located outside the US, a recent notice of proposed rulemaking to extend US security review over some forms of outbound investment could provide the basis to catalog and potentially block investment.⁹⁸Targeted sanctions are another option for limiting investors’ behavior via designating spyware vendors, blocking financial transactions, or designating investors themselves if their actions fall under the scope of spyware-related sanctions authorities. The use of unilateral sanctions is an active and much-debated topic in both the cybersecurity sector and the wider national security policymaking landscape and this report offers a more fully articulated set of policy recommendations based on these trends and the totality of this dataset in the next section.

Policy Recommendations

The 2024 Report on the Cybersecurity Posture of the United States from the Office of the National Cyber Director (ONCD) lists the growing market of “sophisticated and invasive cyber-surveillance tools” as one of the trends driving change in the United States’ cyber strategic environment in 2023.⁹⁹ The UK and French governments have made the proliferation and irresponsible use of commercial cyber intrusion capabilities an important and ongoing policy activity, most notably by leading the deliberately multilateral Pall Mall Process on Cyber Intrusion Capabilities.¹⁰⁰ A European Parliamentary committee (the PEGA committee) highlighted the importance of the spyware market as a topic for policymaking and its implications for technology policy, human rights, and national security across the EU’s complex network of delegated powers. These efforts are part of a degree of sustained attention on spyware not seen in the previous decade. The authors are encouraged to note the effective adoption of some of their previous recommendations.¹⁰¹However, much remains to be done.

The final section of this report presents a set of policy recommendations to further advance these efforts. Not every action identified here is suitable for every state. The United States has outsized authorities and resources and sits in a unique position in the international financial system. Bearing that in mind, however, these recommendations envision a necessary cooperative international approach, conscious of the clear trend of the spyware market’s global infrastructure. While a well-regulated and more transparent market will not entirely prevent proliferation, it can be better channeled, subjected to controls, and made less opaque and harmful, as done with other markets of dual-use goods.

This project is focused on the spyware market, with the goal of fostering greater transparency across the market, limiting jurisdictional arbitrage, and more effectively scrutinizing supplier and investor relationships. These recommendations are built on a sample of this market, not a final and definitive record of all entities and relationships. Achieving these policy goals would address significant opportunities to limit the risks and harms stemming from the proliferation of spyware. These include more granular and effective policies on vendor behavior, a robust investor due diligence regime, information-rich government risk assessments prior to procurement, and credible legal support for greater long-term transparency in the market.

Figure 13: Mapping spyware trends to policy recommendations

1. Mandate “Know Your Vendor” Requirements

If the defining characteristic of spyware acquisition should, in theory, be an exercise of due diligence, then a marked gap in current policy is the ability to exercise that diligence in the face of shifting vendor identities and supply chains. There is, however, a practical solution: employing “Know Your Vendor” (KYV) requirements. Building on previous recommendations made by the Cyber Statecraft Initiative, the United States and, at a minimum, the sixteen additional signatories to the Joint Statement, should enforce KYV requirements that spyware vendors disclose supplier and investor relationships.¹⁰²This is a credible step toward better information about the segments of the spyware market with which these governments might do business. Such a KYV requirement, implemented consistently across these states, would present a united front to many of the vendors covered in this report who claim to work only with “government” and “Western government” clients. This would also mitigate the potential impact of individual governments fearing vendors would turn down their business.

KYV would create a more consistent reporting environment on the spyware market in these states, providing government clients with the ability to check where their prospective supply chain might include firms on restricted entity lists before awarding contracts. With straightforward information sharing, KYV would also enable long-term efforts to reduce government dollars flowing to high-risk suppliers or vendors. A more effective version of these requirements could mandate disclosure of firms further down the supply chain (suppliers to the suppliers of a vendor).

The United States could set KYV requirements through the US Federal Acquisition Regulatory Council, which would require an update to the Federal Acquisition Regulation (FAR) and the Defense Federal Acquisition Regulation Supplement (DFARS) to mandate that any company submitting a bid for a government contract for cyber operations to disclose a list of their vendors and suppliers, investors, and any parent corporate or holding entity. A notice and comment period for such a requirement would likely see vendors request more targeted disclosure requirements for larger conglomerate firms and mechanisms to scope KYV to spyware business units in these firms would be appropriate. KYV would also complement and be strengthened by more effective beneficial ownership requirements.

2. Improve Government-Run Corporate Registries

Similar to KYV, government-run corporate registries are a resource for due diligence and accountability. These registries would play a significant role in a more assertive policy regime that addresses the cross-border movement of, or investment in, spyware vendors. These registries would also be an important source of information for due diligence by potential investors as well as provide improved visibility into business entities operating within respective jurisdictions

For corporate registries to become sources of truth on business and financial structure and histories of vendors, they must be comprehensive, openly accessible to the public, and have verified information. However, currently, the information in corporate registries varies from country to country. For example, the corporate registry of the Czechia is comprehensive and contains information about the different names a company has used since its inception, its history of investment by various investors, as well as the individuals who held senior executive offices and their tenures.¹⁰³ In contrast, the registries in India and Israel provide only basic information about entities such as the legal name of the corporation, address, date of incorporation, and registration number.“¹⁰⁴In the United States, every state maintains a separate corporate registry of the entities incorporated in their jurisdiction.

2.1 Expand the Minimum Scope of Data Captured by Registries

National regulations should determine requirements about the categories and corresponding details present in their corporate registries.¹⁰⁵ They should include basic company information (name, registration number, payment ID, address, contact details, and date of registration), ownership details (senior executives, management board, and actual beneficiaries), the number of employees, financial information (balance sheet, cash flow, income statements, and investors), history of name changes, and the legal status of activity (liquidated, active, or bankrupt). This information serves as a bare minimum but could be expanded to include a history of mergers and acquisitions, legal actions against the firm, and active export licenses.

In the United States, this could be accomplished through the National Association of Secretaries of State (NASS) providing guidance to each of the fifty US states. Alternatively, given some risk of a race to the bottom between US states eager to attract corporate activity, the IRS could publish this data where it is collected for Federal purposes.¹⁰⁶

Outside the US, given the global nature of the spyware market, there is merit in improving corporate registries, especially for the seventeen countries signed on to the Joint Statement on Efforts to Counter the Proliferation and Misuse of Commercial Spyware. As the Joint Statement evolves, the Department of State’s Bureau of Cyberspace and Digital Policy should consider developing subject-specific working groups—including one on corporate registries and associated information—to bolster harmonization and information sharing across signatories. These countries should seek to work towards seamlessly sharing this registry data where it is not made public. Regularly cross-referencing information on a vendor, especially ones with branches or subsidiaries in multiple countries, can be beneficial in avoiding jurisdictional hopping and arbitrage.

2.2 Expand Beneficial Ownership Identification

In January 2024, the US Department of the Treasury unveiled its Beneficial Ownership Program (BOP), seeking to improve corporate filings on the “persons who ultimately own or control the company.”¹⁰⁷ Most countries, however, do not have reporting requirements on beneficial ownership, and many that do have insufficient standards.¹⁰⁸ Better recognition of the beneficial owners behind spyware vendors and, eventually, suppliers, would provide a strong counter to many of the identified trends in this report.

Further improvements can be made to BOPs worldwide. Analysis of beneficial ownership registries of G7 countries Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States (all of whom, except for Italy, are signatories of the Joint Statement) showed that while these countries maintain beneficial ownership registries that require companies to report individuals who own twenty-five percent or more of the shares and voting rights or exert significant control over the management of the company, most do not require any identification document to be submitted by such individuals for verification.¹⁰⁹

While the US entity that is responsible for beneficial ownership registration, the Department of Treasury’s Financial Crimes Enforcement Network (FINCEN), requires entities to provide an image of an identification document for every beneficial owner, the United Kingdom’s Companies House register takes no steps to verify name or address data provided by applicants.¹¹⁰ As highlighted by the capital crosses borders trend, the United States and the United Kingdom are two (of several) concentrated hosts of investors in the spyware market in the sample provided by this dataset. It is especially important for these jurisdictions to improve their methods for identifying beneficial owners.

Other signatories of the Joint Statement do not have BOPs. Transparency Data’s study of European business registers reveals that out of the eleven central and eastern European countries that they analyzed, only three required companies to provide information about actual beneficiary owners. Finland, Sweden, and Switzerland (all signatories on the Joint Statement) do not even require information about the founders or owners of companies.¹¹¹

Beneficial ownership transparency is endorsed and monitored by several international organizations, such as the Financial Action Task Force (FATF) and the OECD’s Global Forum on Transparency and Exchange of Information for Tax Purposes.¹¹² The seventeen countries that are signatories of the Joint Statement without BOP programs or reporting requirements on founders or owners of companies should enact reporting requirements within their respective jurisdictions which harmonize with both US approaches and global best practices.

2.3 Make Government-Run Corporate Registry Data Public

A final improvement to these registries would be to provide universal public access to their data, in all jurisdictions. OpenCorporates’ report, “The Closed World of Company Data,” scores countries on the openness, depth, and accessibility of national corporate data records.¹¹³ While the average score is only twenty-two percent of the maximum, several major states—Spain, Greece, and Brazil—scored 0, which means that their corporate registers cannot even be searched without some form of payment.¹¹⁴ The United Kingdom’s scored highest while the United States was only a few ticks above the average.¹¹⁵ The poor quality of these records hampers due diligence efforts by many actors. For example, high-quality and publicly accessible corporate databases would provide raw informational material for a significantly enhanced investor due diligence regime in the spyware market. It would also help to level the playing field between regulators in different jurisdictions. This would enable cross-border collaboration in regulating the behavior of spyware vendors, suppliers, and investors. This would also allow entities subject to customer due diligence (e.g. banks, notaries, corporate service providers) to improve their verification processes and report discrepancies.

3. Enrich, Audit, and Publish Export Licenses

Export licensing requirements are a mechanism for governments to limit the sale and use of certain products and services outside of their borders. Export licenses are a complex domain characterized by significant inter-state variability in standards, covered goods, and application of broader tests of public interest, such as human rights considerations.¹¹⁶ Indeed, authorities may deprioritize human rights risks if countervailing considerations such as industry growth or perceived geopolitical influence weigh in favor of license approval.¹¹⁷ The Joint Statement commits its signatories to implement export controls on spyware technology in accordance with their respective laws and regulations.

Certain spyware vendors, like NSO Group, have publicly capitalized on the fact that their exports are licensed by government agencies as an indication of their lawfulness.¹¹⁸ Decoupling licensing decisions made in deliberate furtherance of geopolitical goals over those dominated by commercial considerations is a tricky and ongoing research question.¹¹⁹ As noted by the UN Special Rapporteur on Freedom of Opinion and Expression, the global export control framework and its national implementation in areas where NSO Group operates are inadequate for regulating surveillance technology or accounting for human rights impacts.¹²⁰ The result is that while NSO Group’s exports are indeed “licensed,” these and those of other vendors could still present a grave risk to human rights, especially in jurisdictions where the legal framework governing the use of its product is minimal or even nonexistent.

Export licensing regimes can act as a legal mechanism to collect vendor records and some limited activity data. This recommendation suggests strengthening them for that purpose in three ways.

First, export control licenses for spyware and closely related services should include the names of all employees whose work has a material impact on the development of the product subject to the export license. This is important information for policymakers on two grounds. First, these employees are tied to a specific product or spyware service in a semi-public record in perpetuity. This may have some deterrent effect in choosing to work with these registered vendors and stops short of lifetime bans or similar sanctions. The record serves as an indicator of behavior at a certain point in time but does not constitute a lifetime scarlet letter. Second, these companies are tied to those employees, also at a point in time. The sample of the spyware market captured in this dataset has shown vendor names and labels to be a fluid construct and policies should be focused on learning, and potentially shaping, the behavior of individuals in this market more directly. Should a vendor seek to shift jurisdiction and apply for appropriate export licenses for the same product in their new domicile, embedded “critical employee” information in these licenses would ensure a paper trail.

The determination of what constitutes material impact, as well as defining covered products beyond the language in the Wassenaar Arrangement, would need to be determined by each county at the national level. In the US, this should come from a Bureau of Industry and Security (BIS) policy guidance. Under the Department of Commerce, BIS is already “actively engaged in formulating, coordinating, and implementing various export controls to counter the use of items subject to the Export Administration Regulations (EAR) that could enable human rights abuses or repression of democracy throughout the world”.¹²¹ More widely, this would inform a working group under the continuing Joint Statement mechanism as the seventeen signatories seek to harmonize their definitions and enforcement of this additional requirement.

Second, to ensure export controls are effective domestically, governments should build mandatory and regular audits into licensure practices along with punishments for non-compliance. In the United States, the BIS is responsible for regulating the export of dual-use products and services under EAR, including the Commerce Control List of Dual-Use Items (CCL). BIS’s Export Compliance Guidelines contain a section on audits with broad guidelines for auditors, but poor execution or a lack of audits is a recurring barrier in the implementation of EAR.¹²²

An export license for software should also include an explicitly time-bound permission to export. The concept of “continuous monitoring” is an approach to security and compliance in cybersecurity that acknowledges software cannot be evaluated and “signed off” on at a single point in time, but rather needs to be continuously tracked throughout its operation. The same is true of the spyware vendor business model and regular audits of these licenses would accomplish much the same ends. The auditing process, conducted by the licensing authority with appropriate specialized support, would enhance transparency in the export control process and allow licenses to be revoked in the face of evidence of abuse or misuse of spyware vendors’ products. An effective audit would scrutinize all aspects of an export licensing process including application procedures, decision-making criteria, approval processes, monitoring mechanisms, and compliance enforcement. In this way, any discrepancies or red flags discovered could be disclosed to partner agencies to execute audit recommendations.¹²³

Third, these audit reports and the original export licenses should also be made accessible to the public by the national licensing authority; this would be BIS in the case of the US. Reasonable redaction of personally identifiable or business-sensitive information could be made but this should be weighed heavily against the significant public interest in greater transparency in the activities of spyware vendors. Public export license records would complement largely private KYV data and allow for the broader research and civil society community to fulfill an important role as external accountability mechanisms.

Export controls are, at best, a marginal utility in regulating the spyware market. Their focus on transactions emphasizes one of the least regulable steps in the spyware supply chain and to expect licensing to address myriad end uses or all facets of vendor behavior would be wildly optimistic. Instead, this recommendation proposes to take export licensing for the marginal benefit it might offer policymakers and deriving additional value from improved transparency—without leaving export controls as the sole, or even most critical, line of defense against the risks of spyware.

4. Limit Jurisdictional Arbitrage by Vendors

Imposing policy outside of a state’s jurisdiction is challenging and presents opportunities for spyware vendors and others seeking to elude regulatory controls. This extraterritoriality is often exploited by spyware vendors to engage in jurisdictional arbitrage and take advantage of inconsistencies across governments. This recommendation outlines several steps to limit this arbitrage, focusing on raising the barrier for vendors associated with an export license for electronic surveillance technologies, including spyware, to leave a single jurisdiction and mandate reporting of new branches and subsidiaries. To keep things simple, this recommendation defines surveillance technologies using the State Department’s language from the Joint Statement: “Technologies used for surveillance can refer to products or services that can be used to detect, monitor, intercept, collect, exploit, preserve, process, analyze, invasively observe, and/or retain sensitive data, personally identifying information, including biomarkers, or communications concerning individuals or groups.”¹²⁴

Jurisdictional arbitrage by spyware vendors undermines the rule of law, efficiency of regulatory supervision, and market integrity, and can trigger regulatory competition amongst different jurisdictions where states adopt low-standard regulatory requirements to attract business and investments.¹²⁵For example, the European Union, theoretically, has strong regulations against spyware vendors including Council Regulation (EC) No. 2021/821 which ensures regulatory consistency across EU member states.¹²⁶However, the bloc faces fragmentation when it comes to implementation with countries like Bulgaria, Cyprus, Greece, Italy, Malta, and Hungary demonstrating highly variable political commitment to or institutional capacity for strict export controls.¹²⁷Intellexa is a prime example of a vendor that established new subsidiaries in these countries to take advantage of jurisdictional arbitrage.¹²⁸

To address this problem, policymakers should first make it more challenging and costly for a vendor with an export license to exit a jurisdiction. This should include mandating the public disclosure of any subsidiary or branch openings or closures by a vendor that has been granted a license to export spyware. Second, policymakers with existing authorities to regulate inbound investment (such as the Committee on Foreign Investment in the United States–CFIUS) should automatically review transactions impacting the ownership structure of domestic spyware vendors in any way. Vendors that fail to flag such transactions (such information may come to light in mandated KYV disclosures, for instance) should be barred from participating in government acquisitions and/or have their export licenses suspended for some period.

Consider the example of the disclosure requirements of the banks which are members of the Federal Reserve System in the United States.¹²⁹These banks are required by the Federal Deposit Insurance Act to seek approval of and disclose to the Federal Reserve (as well as the public) any instances of openings, closures, or mergers with another bank, including the closure of any bank branches as a result of mergers and acquisitions.¹³⁰ This requirement ensures that the Federal Reserve can monitor and supervise the geographical footprint and operational changes of banks across different states and regions.¹³¹ In the context of the spyware market, a notification requirement would enhance the transparency of the market and help put regulators in different jurisdictions on a more equal footing. It would also allow key stakeholders, including civil society organizations, to have visibility into the operations of spyware vendors and their compliance with regulatory requirements.

Third, harmonizing these disclosure requirements should be the subject of another working group under the Joint Statement or a similar collaborative mechanism. Effective barriers to market exit by these spyware vendors would help to improve the potential influence of domestic policies, including those across the EU, on these firms. Consistency in those barriers would reduce the incentives toward arbitrage.

5. Provide greater protection against Strategic Lawsuits Against Public Participation (SLAPP)

One of the abiding trends of this research and the work of the broader analytic community examining spyware is the tremendous importance of open reporting. There is no substitute for applied public policy analysis in this space but the relationship between journalism and research is deeply symbiotic. A disturbing recent trend threatens to undermine this reporting as a handful of spyware vendors deploy Strategic Lawsuits Against Public Participation (SLAPP).¹³²

In 2022, the news outlet Reuters was sued for defamation against a vendor’s parent company profiled in this report, Appin Technologies and Appin Security Group. As a result, Reuters removed an investigation into the group’s activities from its website.¹³³ This lawsuit sets a dangerous precedent for journalists and researchers alike who offer much-needed transparency into an already opaque market.

To address the harms and frequency of SLAPP suits more generally, the European Commission established a set of rules in May of 2024 that provide heightened protections for speech on matters of “public interest.”¹³⁴ The essential elements of these rules would be a suitable starting point for comparable policies in the United States and elsewhere, including:

Accelerated treatment of issues raised under these heightened protections.

The possibility for early dismissal of “claims against public participation” which are determined to be “manifestly unfounded” at “the earliest possible stage in the proceedings, in accordance with national law.”

Provision for the recovery of all costs of the proceedings by defendants and the potential for application of “effective, proportionate and dissuasive penalties or other equally effective appropriate measures, including the payment of compensation for damage or the publication of the court decision” on the party initiating the action.¹³⁵

It would also be welcome for states that host the victims of SLAPP suits to raise the issue through existing diplomatic channels with states hosting parties initiating these suits. In practical terms, the State Department should address the impact of the Appin suit against Reuters with the Indian and UK governments (as the domicile of the claimant vendor and the court of jurisdiction respectively). This does little to impact the suit directly but shows awareness by the US government and may raise the costs of action by firms in these countries.

Areas for Future Work

These recommendations focus on achieving greater transparency across the spyware market, limiting jurisdictional arbitrage by vendors, and more effectively scrutinizing supplier and investor relationships with those vendors. They do not address the full range of issues that urgently need greater attention and resolution in the proliferation of spyware. This section addresses these opportunities for future work and concludes with a call for consensus and action by at least a small group of states to advance these and related policies.

Bringing the brokers back in: There is insufficient coverage of supplier firms in this dataset relative to the number of spyware vendors. These firms, some of which might be categorized as exploit “brokers,” are important to the discussion of how to reduce the harms associated with spyware, but they are less widely reported on and not systematically understood. These firms traffic in information, and to some degree talent, whose product is not intrinsically malicious (and which has been subject to poorly conceived controls in the past). At the same time, the activities of these suppliers and brokers are a critical wedge between advocates for more effective spyware policy rooted in national security concerns and those advocating from a human rights-centered perspective (acknowledging a degree of overlap between the two). Better information on the diversity of locations, activities, and organization of these firms would benefit an otherwise opaque segment of this market whose activities extend far beyond spyware. The length of supply chains across the spyware market, and the number of steps between suppliers for high-risk vendors as well as governments otherwise practicing adequate due diligence against abuse, could be a substantial driver of unaddressed risk and merits further investigation.

Spyware vendors or suppliers partnered with major tech firms: There are several instances where spyware vendors or suppliers have formed partnerships with conventional technology firms. Positive Technologies may be the most notable, having previously been a member of Microsoft’s Active Protections Program (MAPP) and publicly advertising its work with Samsung.¹³⁶ The structure of these relationships may be tied to vulnerability disclosure, underlining the complex role played by vulnerability discovery and exploitation in both offensive and defensive activities.

A “who’s who” for spyware investors: The incentives for different kinds of investors (for example, venture capital firms vs. private equity) are clear in conventional markets, but much less so in the spyware sector. Both types of entities appear in this dataset. Designing a more robust due diligence regime for investors in these vendors would benefit from a more precise understanding of the motivations of different types of investors for entering the market.

The customer might often be wrong: This project has not yet covered the range of government customers to which this market largely caters. The behavior of these agencies, parties, offices, and bureaus should be of significant interest as they are the ultimate source of demand that shapes the spyware market. States willing to take affirmative action against the spyware market may also quickly find they have better existing tools to shape the behavior of their allies and partners, integrating spyware into wider defense assistance, trade, and legal cooperation agendas. Cataloging customers, their relationships with specific firms and associated supply chains, and the timing of these relationships is a fruitful area for future work. How those customer relationships form and their portability between spyware vendors is also worth continued analysis.

A role for technology companies: Technology firms have a role to play in shaping the spyware market, if for no other reason than they may be supplying Software-as-a-Service (SaaS) and other technologies to spyware-related firms. Research from Amnesty International and others in 2021 established the NSO Group was using AWS products as part of the command-and-control infrastructure for its spyware product and even earlier reporting from Motherboard pointed to NSO delivering its product from an Amazon IP address.¹³⁷Discovery as part of an ongoing lawsuit between Meta and NSO established that NSO became an AWS customer beginning in 2018.¹³⁸

Technology companies might play at least two roles in further shaping the spyware market. First is executing due diligence on technology sales, especially export-controlled hardware and cloud services over a certain dollar threshold, to determine if the customer is an identified participant in the spyware market. Those identified customers should be subject to scrutiny and possible removal but early identification would help cloud vendors avoid discovering via the media that their services are being used in the development or deployment of spyware. Second is major technology firms, perhaps only cloud service providers, developing a common code of conduct as to how they sell services to participants in the spyware market and under what conditions they might limit or refuse sales. The development of both this review policy and the code of conduct are prospects for future work. In the meantime, major technology companies like Meta, Apple, Google, and Microsoft remain some of the only entities with both the standing and the resources to conduct sustained legal action against vendors selling products that create significant risk or harm.

Whistleblowers: While open reporting and journalism are important sources of transparency in this market and deserve heightened protection, the status of whistleblowers from spyware vendors or customers is uncertain. Consolidated guidance as to the human rights or procedural responsibilities of spyware vendors and their customers, accounting for jurisdiction, would help to clarify under what conditions an employee or government official might act with adequate legal protections. Specifying the kinds of legal or ethical expectations of vendors and customers, as well as clarifying protections both against retaliation and to navigate limits on publishing proprietary information, at a national level, would help strengthen an important channel of information on the use and abuse of these products.

Clarifying de-listing procedures: An important feature of an effective sanctions regime is a transparent de-listing procedure.¹³⁹ Given the purpose of imposing sanctions is to cause behavior change in the sanctioned entity or individual, sanctioned entities and individuals should be able to request de-listing once they have demonstrably changed their practices. This is important for ensuring the legitimacy and credibility of sanctions regimes. Hence, governments and multilateral organizations should clearly specify the process and conditions that result in the de-listing from their sanctions list and both researchers and advocates can help map out these processes.

What comes next: How do these and other proposed changes impact the shape and jurisdictional concentration of the spyware market? One of the risks for more assertive regulation of spyware development, sales, and use is vendors, suppliers, or other entities moving to jurisdictions outside the reach of those states using policy to shape this market. The suggestion is not unwarranted in no small part because of the jurisdictional arbitrage already observed in this market. More work is needed to understand what policies might pose the greatest likelihood of this market shift or its segmentation into multiple tiers. This dataset (a sample of the market) shows that even vendors in “hard to reach” jurisdictions with local governments unwilling to regulate their behavior still rely on foreign capital and suppliers. These investors and suppliers are often based in states with a demonstrated willingness to enforce existing policies on spyware. Therein lies the opportunity for change.

Conclusion

There is a certain macabre humor in the lengths that spyware firms go to in obscuring their true nature and purpose, disguising themselves as mythical beasts in an obscured global market. The purpose of this report is primarily to demystify—or demythify—the global spyware market, moving beyond coverage of individual firms to unveil a network of relationships between spyware vendors, suppliers, and investors across forty-two different countries. This data is only a sample but it evidences several trends, including cross-border financial support, shifting vendor identities, and a pattern of jurisdictional arbitrage which would undermine discrete national-level efforts to reshape this market. There is more to be discovered about this market and the authors’ sincere hope is that this project provides support to many other researchers, analysts, and advocates.

Policies that work to regulate and influence the spyware market, which are coordinated amongst at least a small set of countries, have better prospects to reduce the harms and risks posed by spyware. Policymakers who succeed at improving transparency in this market, raising barriers to vendor reorganization and reincorporation, and applying greater scrutiny on supplier and investor relationships will directly confront critical drivers of spyware’s proliferation and abuse. The recommendations presented in this report address these three priorities while laying the groundwork for granular controls on transactions and policies based on distinctions in the legitimate use of spyware.

Most available evidence suggests that spyware sales are a present reality and likely to continue. Proliferation heedless of its potential human rights harms and national security risks, however, is not a stable status quo. Little of the present market for spyware is regulated or governed well enough to address these harms and risks. In some areas, there is a pressing need for additional research but in many others, the initiative sits with policymakers. Nascent steps by a handful of countries demonstrate that a more vigorous approach to shape the behavior of spyware vendors, their supply chain, and their investors is possible. Where such progress has been stymied by a lack of systematic data on this market, Mythical Beasts offers a contribution. However, much more remains to be done.

Acknowledgments

Mythical Beasts was made possible with the support of Microsoft and the UK’s National Cyber Security Centre, we extend sincere thanks to the teams in Redmond and London for their commitment to this work. This project owes a debt of gratitude to the security research community, in particular Sophia d’Antoine, and contributors both inside and out of the Cyber Statecraft networkFor early conversations that helped shape this research thank you to Kirsten Hazelrig, James Shires, and several others who wish to remain anonymous. For peer review of this report, the dataset, and its interactive, thank you to Chris Delaney, Kaja Ciglic, John Hering, David Agranovich, Ingrid Dickinson, Emma Schroeder, Jorn Fleck, Clement Lecigne, Salo Aburto, Jessica Dabrowsky, James Batchik, Constantine Stasinopoulos, Jennifer Brody, and Lisandra Novo.

Thank you to Stewart Scott and Alexander Beatty for their contributions which early on shaped the focus of this report and perspectives on its art. Thank you to Nancy Messieh for all the graphics and to the team at Schema Design for the interactive version of this dataset. Thanks and appreciation are due to Andrey Prokopenko for the artwork and for his creative output despite terrific challenges. Major credit to Emma Taylor as well as Jean Le Roux and Sopo Gelava without whose collaboration the dataset and its presentation would not have been possible. Thank you to Natalie McEvoy, Charlette Goth-Sosa, Kristopher Kaliher, and Donald Partyka for review, editing, and production. For feedback on this project as it evolved thank you as well to the attendees of more than half a dozen roundtables and events over the past year and to the members of two different communities for your time, questions, and willingness to engage.

About the Authors

Jen Roberts is an Assistant Director with the Atlantic Council’s Cyber Statecraft Initiative. She primarily works on CSI’s Proliferation of Offensive Cyber Capabilities and Combating Cybercrime work. Jen also helps support the Cyber 9/12 Strategy Challenge and is passionate about how the United States with its allies and partners, especially in the Indo-Pacific, can cooperate in the cyber domain. Jen holds an MA in International Relations and Economics from Johns Hopkins University’s School of Advanced International Studies (SAIS) where she concentrated in Strategic Studies. She also attained her BA in International Studies from American University’s School of International Service.

Trey Herr is assistant professor of Global Security and Policy at American University’s School of International Service and Senior Director of the Atlantic Council’s Cyber Statecraft Initiative. At the Council, the CSI team works at the intersection of cybersecurity and geopolitics across conflict, cloud computing, supply chain policy, and more. At American, Trey’s work focuses on complex interactions between states and non-state groups, especially firms, in cyberspace. Previously, he was a senior security strategist with Microsoft handling cybersecurity policy as well as a fellow with the Belfer Cybersecurity Project at Harvard Kennedy School and a non-resident fellow with the Hoover Institution at Stanford University. He holds a PhD in Political Science and BS in Musical Theatre and Political Science.

Nitansha Bansal is an assistant director with the Cyber Statecraft Initiative (CSI), part of the Atlantic Council Tech Programs. In this role, her research focuses on the proliferation of offensive cyber capabilities, including spyware and its policy implications for human rights and national security, and open source software security. She also supports the capacity building efforts of CSI, and runs the Congressional Cyber and Digital Policy Program. Prior to joining the Council, Nitansha worked with government and think tanks in India on technology policy. She holds a masters in public administration from Columbia University’s School of International and Public Affairs where she concentrated on cybersecurity and business risk, social media policy, and data analysis.

Nancy Messieh is Deputy Director for Visual Communications with the Atlantic Council’s Cyber Statecraft Initiative, where she focuses on visual storytelling and data visualization.

Emma Taylor is a Research Assistant with the School of International Service and a highly interdisciplinary professional pursuing an M.S. in Computer Science and Cybersecurity with previous experience in the technology industry.

Jean Le Roux conducted this research as a research associate for the Sub-Saharan Africa region at the Atlantic Council’s Digital Forensic Research Lab (DFRLab). He is now a senior investigator at Graphika.

Sopo Gelava is a research associate for the Caucasus with the Atlantic Council’s Digital Forensic Research Lab. Prior to the DFRLab, she served as media literacy programs director at Media Development Foundation, leading Georgian think-tank countering disinformation and information operations.

Appendices

Appendix A- Supplier and Vendor Profiles

This appendix summarizes the profiles of suppliers and vendors within the dataset that are included in the analysis of this report. The authors sought to include this information to outline details about all entities within the dataset.

Suppliers

Azimuth Security

In 2010, Mark Dowd and John McDonald founded Azimuth Security, an Australia-based exploit developer and boutique hacking firm.“¹⁴⁰ The company gained notoriety for its role in unlocking the San Bernadino shooter’s iPhone in 2016. In 2018, an American firm, L3 Technologies, now L3Harris, purchased Azimuth Security and Linchpin Labs of Canada.¹⁴¹Today, Azimuth and Linchpin Labs operate under the brand name “Trenchant.” The Trenchant group of companies operates across three jurisdictions: United Kingdom, Canada, and Australia. L3Harris Trenchant Canada Inc. (Canada), and Australia.¹⁴²In the past, Azimuth supposedly restricted sales to members of the Five Eyes intelligence alliance—Australia, Canada, New Zealand, the United Kingdom, and the United States.¹⁴³ Trenchant does not currently face any restrictions to its exploit sales or business operations.

Blue Oceans Technologies

Blue Ocean Technologies is an Israeli supplier that was incorporated in 2015¹⁴⁴ by retired Brigadier General Rami Ben Efraim and Lieutenant Colonel Ron Tira.¹⁴⁵An Israeli newspaper, The Globes, reported that Blue Ocean Technologies is an exception in the Israeli spyware market since it was established as part of a deal between an East Asian country and the founders of the firm.¹⁴⁶Intelligence Online claims that the East Asian country is Singapore, and Blue Ocean Technologies received two export licenses from the Israeli Defense Ministry to provide the Singaporean Ministry of Defense with a team of vulnerability researchers to weaponize Singapore’s cyber tools.¹⁴⁷

Brigadier General Rami Ben Efraim, through his strategic consulting firm Lee and Rami Ben-Efraim Ltd. (also known as BNF Group), holds options in Blue Ocean Technologies.

Computer Security Initiative Consultancy PTE Ltd. (COSEINC)

Founded in 2004 in Singapore,¹⁴⁸ Computer Security Initiative Consultancy PTE Ltd. (known widely as COSEINC) is known for distributing exploits without control and known to host pwn0rama—its own cyber vulnerability acquisition program¹⁴⁹—and is classified as a supplier within this dataset. The Bureau of Industry and Security (BIS) of the US Department of Commerce added COSEINC to its Entity List for Malicious Cyber Activities in November 2021 based on a BIS determination that the vendor “traffic[s] in cyber tools used to gain unauthorized access to information systems, threatening the privacy and security of individuals and organizations worldwide.”¹⁵⁰

COSEINC was founded by Thomas Lim, who is known for organizing a security conference, SyScan, until it was sold to Chinese technology firm Qihoo 360, another sanctioned entity.¹⁵¹ In 2015, WikiLeaks exposed Lim’s attempt to sell hacking tools to Italian Spyware vendor Hacking Team srl,¹⁵²thereby, hinting at a possible connection between COSEINC and Hacking Team srl.¹⁵³In 2022, the company became inactive.¹⁵⁴

Crowdfense Technological Project Management – Sole Proprietorship LLC

Founded in 2017 in the United Arab Emirates (UAE), Crowdfense Limited buys, develops, and sells zero-day exploits that target a variety of platforms. In 2018, Crowdfense Limited launched its first bug bounty program with a $10 million budget.¹⁵⁵Since then, the company has continued to grow its bug-bounty budget year over year as it expands the scope of its “interest” areas. According to the UAE business registry, Crowdfense Limited dissolved in 2023 and a new entity named Crowdfense Technological Project Management – Sole Proprietorship LLC was registered. In 2024, Crowdfense Technological Project Management-Sole Proprietorship LLC boasted a $30 million budget that now includes exploit acquisitions related to “Enterprise Software, WiFi/Baseband and Messengers.”¹⁵⁶The company maintains an unknown number of offices in Abu Dhabi¹⁵⁷ and some reporting indicates it receives financial backing from the governments of the UAE and Saudi Arabia.¹⁵⁸

Dataflow Security s.r.l.

DataFlow Security s.r.l. (AKA DFSEC) was founded in 2022 by Ofer Cohen.¹⁵⁹Based in Italy, the company specializes in vulnerability research and exploit development.¹⁶⁰ This report classifies DFSEC as a supplier due to its development, optimization, and sale of exploits. DFSEC’s internal client website contains a catalog of exploits for purchase. In 2022, Dataflow Security Spain SL was established in Spain.¹⁶¹ In the same year, Dataflow Forensics was established as a sister company to DFSEC focused on defensive cybersecurity operations.¹⁶²DFSEC acquired a majority stake in Random Research, an Israeli company also founded by Ofer Cohen.¹⁶³ At this time, there is little available information concerning DFSEC funding. However, per the official Spanish corporate gazette, the sole shareholder of Dataflow Security Spain SL is Dataflow Security s.r.l., and while there has been no update to the company’s shareholders since its incorporation its share capital increased from 3,000 Euros to 153,000 Euros on June 28, 2024.¹⁶⁴This commonly indicates a new investment and/or a new shareholder. However, limited companies are not required to declare shareholders in the Spanish public gazette. This group of firms has not faced any significant roadblocks to business operations.

PARS Defense

Registered in Turkey in 2021, PARS Defense was founded by Ibraham Baliç, an individual who has been operating as a “vulnerability specialist” since 2010.¹⁶⁵PARS Defense specializes in detecting vulnerabilities and operating codes on mobile systems and is coded as a supplier within the dataset. Google identified two vulnerabilities attributed to PARS defense that were present in iOS.¹⁶⁶

No information was found on PARS Defense subsidiaries, partners, holding companies, or investors.

Protect Electronic Systems LLC

Founded in 2016, Protect Electronic Systems LLC, also known as Protect and Protected AE, is a supplier based in the United Arab Emirates.¹⁶⁷The company was reportedly founded from what remained of DarkMatter’s zero-day exploit.¹⁶⁸ More recently, Protect Electronic Systems received attention due to its “special relationship” with Variston IT, a vendor tracked in this report.¹⁶⁹Protect Electronic Systems built upon Variston spyware’s “framework and infrastructure” to create a polished product to sell directly to brokers and governments.¹⁷⁰ At this time, little is known regarding Protect Electronic Systems’ investor base; however, some sources indicate the company may receive state funding.

RebSec Solutions

RebSec Solutions was incorporated in 2012 by Vishvadeep Singh in India and is classified as a supplier within this dataset. It was not possible to identify any institutional or angel investors in RebSec Solutions but the quality of data in open reporting on this firm is limited.¹⁷¹

Zerodium LLC

In 2015, Chaouki Bekrar founded Zerodium LLC in the United States. Bekrar previously founded and led Vupen, a French zero-day exploit vendor. Vupen clients reportedly included “vetted” NATO government agencies, specifically the US National Security Agency (NSA).¹⁷² After Vupen dissolved in 2015, Zerodium LLC emerged to provide identical services in the zero-day exploit industry.¹⁷³Vista Incorporations Limited is Zerodium LLC’s registered agent in Delaware.¹⁷⁴Amidst a market typically shrouded in financial mystery, Zerodium LLC was one of the first firms to put out ads detailing desired exploit specifications with corresponding prices.¹⁷⁵ Other companies, including Russia’s OpZero, Have followed suit and adopted similar public marketing strategies.¹⁷⁶ Currently, Zerodium LCC is a privately held venture capital-backed company; however, little information exists concerning the company’s investor base.¹⁷⁷

Vendors

Aglaya Scientific Aerospace Technology Systems Private Limited

Aglaya Scientific Aerospace Technology Systems Private Limited (Aglaya) was founded in 2014 in India by Ankur Srivastava. In addition to spyware services, Aglaya markets itself as a zero-day seller and a censorship-as-a-service company specializing in online trolling and disinformation.¹⁷⁸Offering to run its buyer’s spyware operations for 2,500 euros per day and 600 euros for disinformation campaigns, Agalya positions itself in an interesting space in this market as it sells to non-government entities.¹⁷⁹

The financial structure of the company is also entirely based in India, with holding companies becoming inactive in 2021, alongside the vendor itself. Aglaya was included in this research report to highlight the market for spyware outside of corporate-to-government sales, as well as the wide range of product types offered by full-service vendors, which can include not only the spyware product itself but command and control package offerings.

Cognyte Software Ltd.

Cognyte Software Ltd. was established as an independent company in 2020, registered in Israel, after the US-based Verint Systems Inc. separated its customer engagement business from its cyber intelligence business due to shareholder pressure. As a result of the separation, Cognyte Software Ltd. focuses on the “security analytics software market.”¹⁸⁰ Its CEO is Elad Sharon and has subsidiaries in India, Brazil, Bulgaria, Canada, Delaware USA, Mexico, the United Kingdom, Israel, Taiwan, Thailand, Germany, Cyprus, the Netherlands, and Romania. Cognyte is a public company that trades its shares on the NASDAQ, with Visa Equity Partners as its largest institutional shareholder.¹⁸¹

Cognyte Software Ltd., however, has a history dating back to 1994 when Verint Systems Inc. was incorporated as Interactive Information Systems Corporation in the United States. The firm developed AudioDisk, a digital surveillance product intended to be used by police and intelligence agencies to record and store wiretap material.¹⁸² Two years later, the company changed its name to Comverse Information Systems Corporation, which later merged with Comverse InfoMedia Systems to create Comverse Infosys.¹⁸³ The US Department of Defense is known to be a customer of Comverse Infosys’ AudioDisk product.¹⁸⁴ After the 9/11 attacks, Comverse Infosys changed its name to Verint Systems Inc. and launched its Initial Public Offer (IPO).¹⁸⁵ In 2013, Verint Systems Inc. was separated from other businesses of Comverse Technology and made a standalone company with Dan Bodner as its CEO.¹⁸⁶ This is the company that eventually became Cognyte Software Ltd.

The company and its subsidiaries have been in controversies since its inception. In 2001, Fox News reported that AudioDisk used by the US government may have been vulnerable, as these systems allegedly had a back door through which the wiretaps could be intercepted by unauthorized parties.¹⁸⁷

In 2006, it was delisted from the NASDAQ due to allegations of being a part of an options backdating scandal.¹⁸⁸ In its 2021 Threat Report on the Surveillance-for-Hire Industry¹⁸⁹, Meta announced that it removed around one hundred Facebook and Instagram accounts linked to Cognyte Software Ltd. The report claimed that Cognyte Software Ltd. “sells access to its platform which enables managing fake accounts across social media platforms to social-engineer people and collect data.”¹⁹⁰ Most recently in 2022, the Norwegian Government Pension Fund Global (GPFG) was recommended by the country’s Council on Ethics to divest from Cognyte Software Ltd. due to the “unacceptable risk that the company is contributing to serious human rights abuses.”¹⁹¹

Cyber Root Risk Advisory Private Limited

CyberRoot Risk Advisory Private Limited (CyberRoot) was founded in India in 2013 by Vijay Singh Bisht, Chiranshu Ahuja, and Vibhor Sharma.¹⁹² In 2013, CyberRoot entered a relationship of “information sharing” with Appin Security Group and BellTroX Infotech Services Private Ltd., both classified as spyware vendors within this dataset. The nature of this information sharing or when it ended is unclear.¹⁹³ CyberRoot, unlike its other Indian vendor counterparties, has a holding company within the United Kingdom named CyberRoot Limited.¹⁹⁴

DataForense s.r.l.

Founded in Italy by Annunziata Cirilloin in 2013,¹⁹⁵ Dataforense s.r.l. is known for its spyware Aretmide/Spyrtacus project. This system allows users to extract data from phones running Android or iOS.¹⁹⁶The company was in liquidation as of 2024 according to the Italian business registry.¹⁹⁷There was little available information about this particular vendor, but it was included in this report and dataset to show the subcluster of vendors emerging in Italy.

DSIRF GmbH

Founded in 2016 in Austria by Stefan Gesselbauer, DSIRF GmbH is known for its spyware SubZero.¹⁹⁸ The company has one known subsidiary, MLS Machine Learning Solutions GmbH, which specializes in the development and implementation of machine learning models.¹⁹⁹ In 2023, the vendor entered liquidation proceedings within the Vienna Court of Commerce.²⁰⁰ It is believed its subsidiary, MLS Machine Learning Solutions, is absorbing the business of DSIRF, and that DSIRF’s lead investor, DSR Decision Supporting Information Forensic, will continue to support the company.²⁰¹

Gamma Group International SAL

First registered in Germany in 2008, Gamma International GmbH, later renamed to FinFisher Labs GmbH in 2012, is the vendor of FinSpy spyware.²⁰² FinFisher Labs GmbH, in collaboration with their supplier, Elaman GmbH, distributed FinSpy to a variety of different government clientele, including entities in Singapore, South Africa, and Turkey, but remained an exclusively German-domiciled vendor.²⁰³In 2022, FinFisher Labs GmbH shut down operations in Germany after legal prosecution.²⁰⁴ Gamma Group’s holding companies are almost entirely in the United Kingdom, British Virgin Islands, and Cyprus and are associated with a single family. These holding companies might be used to filter investment from this family to Gamma Group. Thus, while Gamma Group is no longer operational within Germany, its financial structure and potentially its investment base are operational.

InvaSys a.s.

In 2017, Kyrre Sletsjøe founded InvaSys a.s. in the Czechia.²⁰⁵ The company specializes in mobile phone interception and qualifies as both a funder and a supplier due to its production of spyware tools and its sales of zero-day vulnerabilities.²⁰⁶ Notably, the company’s Kelpie program provides backdoor access to Android and iPhone devices and encrypted messaging applications.²⁰⁷ The company operates out of two offices in the Czech Republic: one in Brno and another in Prague. Founder and CEO Kyrre Sletsjøe also owns and runs Defense System Property Protection, a physical security firm, and YX Systems. Some InVasys employees previously worked at Sletsjøe’s prior company CEPIA Technologies.²⁰⁸From March to August of 2017 Thomas Vestyby Jensen was listed as the sole owner of InVasys Technologies; however, as of 2022, Kyrre Sletsjøe has a ninety-one percent ownership stake in InvaSys. At present the company has not faced any challenges to its operations in the Czechia.²⁰⁹

Leo Impact Security Services

Founded in 2009 by Manish Kumar, Leo Impact Security Services is listed as a vendor in the dataset.²¹⁰ There has been some reporting that this vendor is a direct competitor of Aglaya, another spyware vendor profiled for this report, as they offer similar spyware products.²¹¹ It has one branch based in the Czechia named Leo Impact Security s.r.o. that has been operational since 2010.²¹²

Mollitiam Industries

Mollitiam is a Spanish vendor founded in 2018 by In-Nova and the cybersecurity firm StackOverflow Ltd. It is headed by Santiago Molins Riera, who is the former head of technology of In-Nova.²¹³ Mollitiam is known to develop payloads that can intercept communications and steal cloud-hosted data from infected devices and deploy spyware on Microsoft, Apple, and Google mobile devices and operating systems. It is known for its interception tools Invisible Man and Night Crawler, which are capable of remotely accessing files and location data and covertly turning on a device’s camera and microphone.²¹⁴

Mollitiam has provided services to Spain’s National Intelligence Centre (CNI) and Mando Conjunto de Ciberdefensa (MCCD), the country’s joint cyberspace command.²¹⁵ It receives funding from the Centre for the Development of Industrial Technology (CDTI) which is a public corporation under the Spanish Ministry of Economy and Competitiveness.²¹⁶ The European Union’s Regional Development Fund provided financial support to Mollitiam between 2019 and 2021 for a project worth €650,000 to build a platform to provide new ways to automatically generate intelligence from data extracted from social media platforms and the dark web.²¹⁷ Apart from these government funds, venture capital firms like EASO Ventures, Sabadell Venture Capital, and Torsa Capital have investments in this Spanish firm.²¹⁸

Movia S.p.A.

Movia S.p.A. is an Italian spyware vendor founded in 2003 by Luca Spina. It is known to sponsor ISS World, a global surveillance technology trade.²¹⁹ The company’s spyware product is called Spider and is used by prosecution offices in Italy. In 2022, the company established a subsidiary called Bioss, and Spina launched another company called Vision s.r.l.²²⁰ Movia’s largest investor is known to be Sistema Investimenti.²²¹Movia was exposed by the Italian anti-terrorist and anti-mafia investigative directorate called Direzione Nazionale Antimafia e Antiterrorismo (DNAA).²²²

negg Group s.r.l.

In 2013, Francesco Taccone co-founded negg Group s.r.l. in Italy.²²³By 2017, Kaspersky Lab published a report detailing the invasive capabilities of Skygofree, a spyware tool it attributes to negg Group.²²⁴ Skygofree ties many of its exploitive services, including audio recordings and photo captures on target devices, to the device location.²²⁵ For example, Skygofree allows attackers to turn on audio recording capabilities when they deem that a device has entered a sensitive location such as a meeting or product development site.²²⁶ Furthermore, the spyware tool forces infected devices to connect to attacker-controlled WiFi networks, offering attackers the ability to collect and analyze WiFi traffic. Finally, this spyware tool exploits vulnerabilities within a device’s accessibility services to allow attackers to read encrypted WhatsApp messages. As of 2024, Meta observed the negg Group accounts testing exploit delivery via Facebook and Instagram and consequently removed its accounts from these platforms.²²⁷ The company maintains three offices registered in Italy: two in Rome and one in Reggio Calabria.²²⁸ Between 2020 and 2022 negg International operated in the Netherlands under the ownership of companies with ties to the negg Group co-founder.²²⁹ However, the business relationship between negg Group and negg International remains unknown at this time. In 2014, the Italian Ministry of Economic Development awarded negg Group a digitalization voucher worth 9,872 euros.²³⁰ At the time, such vouchers were intended to support the digital transformations of Italian companies. At present, the negg Group website states that the company “actively seeks” investors.²³¹

Positive Technologies AO

Founded by Yuri Maksimov and Dmitry Maximo in 2002, Positive Technologies AO is a Russian company that was added to the list of entities sanctioned by the US Office of Foreign Assets Control (OFAC) in 2021 on account of its role in the organization of the Positive Hack Days cybersecurity conference. The conference is said to be used by the Russian Federal Security Service (FSB) for recruitment, according to the US Treasury Department. The Bureau of Industry and Security (BIS) of the US Department of Commerce also accused Positive Technologies AO of distributing exploits and added it to the Entity List for malicious cyber activities.²³² The US Department of State announced the vendor was listed based on a determination that it “misuse[s] and traffic[s] cyber tools that are used to gain unauthorized access to information systems in ways that are contrary to the national security or foreign policy of the United States, threatening the privacy and security of individuals and organizations worldwide”.²³³ Positive Technologies AO has a corporate presence in at least six different countries.

According to Intelligence Online, Positive Technologies operates two websites—one for the Russian market and another for the international market. On its website for the international market, the vendor lists Lukoil, Vimpelcom, Sberbank, the South Korean companies Hanwha and Samsung, France’s Societe Generale bank, and the French cybersecurity agency, Agence Nationale de la Sécurité des Systèmes d’Information (ANSSI), as its clients.²³⁴

RCS Labs

Founded in 1992 in Italy, RCS Labs (RCS ETM Sicurezza S.p.A) operates as both an original producer as well as an intermediary seller of spyware.²³⁵ As early as 2012, RCS facilitated the sale of Hacking Team srl products and services, including Hacking Team srl’s Remote Control System (RCS), to government agencies in Bangladesh, Pakistan, and Turkmenistan.²³⁶ In 2022, security researchers at Lookout determined RCS Lab S.p.A created and sold the Hermit spyware.²³⁷

The RCS Group (formerly Aurora Group) owns RCS Labs. In March 2022, another Italian firm that sells spyware amongst other products, Cy4Gate, acquired Aurora Group, including its seven subsidiary companies: RCS ETM Sicurezza S.p.A., RCS LAB GMBH, Tykelab, Azienda Informatica Italiana, Servizi Tattici Informativi Legali, Dars Telecom SL, and Aurora France S.A.S.²³⁸ Cy4Gate is one of Italy’s largest technology companies, is publicly traded, and its primary investors are Elettronica Group and Expert System. According to Cy4Gate’s 2023 financial reporting, RCS Labs remains the most profitable company in the Aurora Group.²³⁹

Variston Information Technology

Variston Information Technology (Variston) was founded in 2018 by Ralf Wegener and Ramanan Jayaraman and is headquartered in Barcelona, Spain.²⁴⁰ Variston is known to develop data collection tools for law enforcement and security solutions in the areas of supervisory control and data acquisition (SCADA) and the Internet of Things (IoT).²⁴¹ Shortly after its incorporation in 2018, Variston IT acquired Truel IT,²⁴² an Italian zero-day vulnerabilities research company. This acquisition helped Variston onboard new researchers and capabilities, including developing its spyware Heliconia.²⁴³ According to reporting from Intelligence Online in May 2024, Variston Information Technology is now effectively defunct.²⁴⁴

Appendix B-Markets Map: Country List (42)

Appendix C- Markets Map: Vendor List (49)

Aglaya Scientific Aerospace Technology Systems Private Limited
Appin Security Group > Approachinfinate Computer and Security Consultancy Grp.
- Adaptive Control Security Global Corporate
BellTroX Infotech Services Private Ltd
Candiru Ltd > DF Associates > Grindavik Solutions Ltd./Greenwick Solutions > Taveta Ltd./Tabatha Ltd > Saito Tech Ltd.
CyberRoot Risk Advisory Private Limited > CyberRoot Software Solutions LTD
Cytrox AD
- Intellexa S.A.
Dataflow Security s.r.l.
DataForense s.r.l
DSIRF GmbH
Equus Technologies > MerlinX Ltd.
Gamma Group International SAL
- Gamma International GmbH > FinFisher Labs Gmbh
Hacking Team srl (Italy) > Memento Labs srl

Hacking Team srl (United States)
Grey Heron (United Kingdom)
Grey Heron (Italy)

Interionet Systems Ltd.
InvaSys a.s.
Leo Impact Security Service PVT Ltd.

Leo Impact Security s.r.o.

Mollitiam Industries
Movia SPA
Negg Group S.R.L

Negg International

NSO Group

L.E.G.D Technologies > Q Cyber Technologies
Westbridge Technologies
Osy Technologies SARL
Q Cyber Technologies SARL

Paragon Solutions
Positive Technologies AO (Russia)

²⁴⁵

Positive Technologies Global Holding Ltd. (United Kingdom)
Positive Technologies Global Solutions Ltd. (United Kingdom)
Positive Technologies S.R.L (Romania)
Positive Technologies S.R.L. (Italy)
Positive Technologies Inc. (United States)
Positive Technologies Czech s.r.o. (Czechia)
Positive Technologies Holding AG (Switzerland)

Quadream Inc.
RCS ETM Sicurezza S.p.A.

RCS MEA DMCC

Variston IT
Verint Systems Inc.

Verint Systems Ltd.
Cognyte Software Ltd. (Israel)

Disclaimer on Sources

More information: All sources for this dataset are open-source and were publicly available at the time of writing. For more on the kinds of data used in this project, see here. We are aware that some links have broken or been removed, and a handful of sources have been taken down in the wake of court orders. We are unable to replace, or host, copyrighted material. For any questions on sourcing, please email cyber@atlanticcouncil.org.

The Cyber Statecraft Initiative, part of the Atlantic Council Tech Programs, works at the nexus of geopolitics and cybersecurity to craft strategies to help shape the conduct of statecraft and to better inform and secure users of technology.

learn more