Fidelity National Information Services, Inc v Montoux Limited (in liq)
[2025] NZHC 2505
•29 August 2025
IN THE HIGH COURT OF NEW ZEALAND WELLINGTON REGISTRY
I TE KŌTI MATUA O AOTEAROA TE WHANGANUI-A-TARA ROHE
CIV-2025-485-242 [2025] NZHC 2505
UNDERthe Copyright Act 1994 and the Fair Trading Act 1986
IN THE MATTER OF copyright infringement, misuse of
confidential information, inducing breach of contract, misleading conduct in trade, and conversion
BETWEENFIDELITY NATIONAL INFORMATION SERVICES, INC
First Plaintiff
FIS CAPITAL MARKETS UK LIMITED
Second Plaintiff
AND MONTOUX LIMITED (IN LIQ)
First Defendant
HEATH LESLIE GAIR
Second Defendant
Hearing: 6 & 7 August 2025
Counsel: B Carey and P Murray for Plaintiffs
J D Haig and M W Snape for Defendants
Judgment: 29 August 2025
Reissued: 9 September 2025
JUDGMENT OF GENDALL J
Introduction
[1] Fidelity National Information Services Inc and FIS Capital Markets UK Limited (together FIS) are global providers of software solutions to customers in
FIDELITY NATIONAL INFORMATION SERVICES, INC v MONTOUX LIMITED (IN LIQ) [2025] NZHC 2505 [29 August 2025]
insurance, healthcare, finance and other fields. One of their products is the software suite ‘Prophet’ which processes data and performs complex mathematics in the field of actuarial science.
[2] Montoux Limited (Montoux) is a New Zealand company that provided actuarial modelling tools and artificial intelligence services. On 29 November 2024, Montoux was placed into liquidation, with Mr Heath Gair (the Liquidator) appointed as liquidator.
[3] On 6 May 2025, FIS filed a statement of claim alleging that Montoux has committed breach of confidence, copyright infringement, breach of contract, conversion and breach of the Fair Trading Act 1986. FIS claims Montoux copied the Prophet source code (including all formulae, variables, values, libraries, structures and logic therein), its object code, documentation and original artistic works including derivations or adaptions (the Copyright Works) and used them to create competing software products. As to FIS’s “Confidential Information”, this is explained in the statement of claim as being (not publicly known) information defined on the basis that FIS’s Copyright Works contain proprietary information and code, including formulae, variables, values, libraries, and definitions.
[4] In the statement of claim, FIS goes on to claim that ‘Montoux is wrongly accessing, copying or otherwise using this Confidential Information without authorisation. And finally, FIS also claims that the Liquidator is threatening to sell or dispose of Montoux’s infringing software.
[5] Following the making of an interim undertaking by the Liquidator not to advertise, publish or sell the Copyright Works and the Confidential Information, FIS now seeks an interim injunction preventing the defendants from:
publishing, copying, communicating, promoting, adapting, using, selling, realising, disposing of, or otherwise dealing with any works that are, or are alleged by the plaintiffs to be, direct or indirect copies of, or that otherwise copy, adapt, incorporate or use any part of the plaintiffs’ Copyright Works or Confidential Information.
[6] FIS also seeks leave to commence this proceeding against Montoux as a company in liquidation, given the Liquidator has not otherwise agreed. Costs on these applications are also sought.
[7] An undertaking as to damages in terms of r 7.54 of the High Court Rules 2016 has been provided by FIS.
[8]The interim injunction application is opposed by Montoux and the Liquidator.
Background
[9] The original version of Prophet was created and launched in the United Kingdom in the late 1980s as a tool to solve complex problems in respect of “actuarial science”.1 These included the pricing of insurance contracts and determining how much capital to hold to address insolvency risk. Prophet was acquired by FIS in 2015. FIS and predecessors-in-interest to Prophet have reportedly invested over USD$100 million into the design, development, implementation and continued maintenance of Prophet. There are more than 10,000 Prophet users in over 80 countries around the world, including New Zealand.
[10] Montoux was incorporated on 20 January 2012, and launched its inaugural actuarial modelling software platform in New Zealand in 2016. In 2019, it began to develop a next generation platform. This was to be an “open system” allowing actuaries to build and edit financial models before inputting data into the financial model to project future outcomes. According to Montoux, their view was that older systems such as Prophet had too many limitations and were not sufficiently user- friendly. Montoux says it saw their next gen platform as an opportunity to compete and provide the best software platform in the actuarial space, enabling actuaries to become significantly more effective through the application of modern technology. In 2022, Montoux acquired new customers for their next gen platform, including Amazon Web Services, and began migrating existing customers from the old platform. By 2024, the platform was ready for deployment.
1 Actuarial science involves using mathematical and statistical models to analyse risk factors and predict financial implications of uncertain events, for the purpose of designing, implementing, pricing, and managing financial products.
[11] Montoux’s revenue for the financial year ended 31 March 2024 was approximately NZD$2.7 million, up from NZD$2.1 million the previous year. As at 7 May 2023, it is claimed that Montoux was valued at $32,500,000.
[12] According to FIS, around June 2024 it discovered and investigated an example of Prophet Code being copied by Montoux’s computer programme known as a “Model Replication Toolkit” in an online video titled “Montoux Valuation Workflow Demo”.
[13] On 1 November 2024, FIS filed proceedings in Delaware, United States of America, against Montoux. Montoux says this was done without warning. It says no usual cease and desist letter was sent from FIS before the proceedings commenced. Montoux contends this resulted in it losing its ability to convert active sales and opportunities, to renew customer contracts, to secure new customers, or to close investor funding rounds. Customers and investors became hesitant to engage with Montoux. As a result, Montoux says it became cashflow insolvent. Accordingly, as noted above, Montoux was placed into liquidation on 29 November 2024.
[14] On 6 December 2024, FIS sent a letter to the Liquidator outlining its concerns. On 17 December 2024, the Liquidator responded dismissing FIS’s claims as “baseless”, “vexatious” and “anti-competitive”.
[15] On 14 January 2025, Montoux sought the protection of US bankruptcy laws to stay the Delaware case. This was declined on 28 January 2025. On 18 February 2025, Montoux’s US subsidiary company filed for bankruptcy protection in the US, staying the Delaware case.
[16] On 25 March 2025, the Liquidator indicated he intended to sell Montoux assets but would not do this until after 1 May 2025. This was to include software that FIS alleged incorporated its copyright-protected and trade secret material.
[17] On 6 May 2025, as noted above, FIS filed the present substantive claim in the High Court, alongside its interim injunction application.
[18] On 16 May 2025, in a hearing before La Hood J, the Liquidator undertook not to advertise, publish or sell the Confidential Information and Copyright Works until a date in July 2025.2
[19] On 3 June 2025, FIS invited Montoux and the Liquidator to discuss an out-of- court solution whereby an independent expert, subject to appropriate confidentiality undertakings, would be granted access to Montoux’s systems. This expert would examine to what extent materials on Montoux’s systems were “tainted” with FIS copyright works or confidential information. They would then work with the parties to securely extract these materials, following which liquidation would resume. This process was estimated to take two months. The Liquidator reportedly did not respond to the proposal. He has deposed in his affidavit that this proposal would effectively freeze Montoux’s assets.
[20] On 12 June 2025, the defendants filed an interlocutory application for further particulars and a proposal for security for costs to be paid.
Competing evidence
[21] There are substantial disputes over whether Montoux has accessed and used copyright works or confidential information from FIS. Competing affidavit evidence from the parties’ respective witnesses is before the Court. For completeness, I now set out brief details of some of that evidence, which at this point I accept is broadly untested.
Martin Sarjeant’s evidence
[22] Martin Sarjeant is the Senior Vice President of Solutions Management, Insurance and Climate Risk at FIS. He deposes that Prophet’s computer code, including the related formulae, variables and definitions, contain FIS’s proprietary and Confidential Information. He says that FIS licenses and grants access to its Confidential Information on a strictly need-to-know and limited basis to individuals under confidentiality restrictions. The measures it takes to protect this information
2 Fidelity National Information Services Inc v Montoux Ltd (in liq) CIV-2025-485-242, 16 May 2025 (Minute of La Hood J) at [7(f)].
include securing physical facilities and digital spaces where it is stored, establishing information security policies, staff training and contracts, using a licence manager, and marking Prophet with an intellectual property notice. Hardware and software controls are used to regulate the number of users of Prophet and how they can access it and its data libraries. Software licences are also employed to control and restrict access to Prophet libraries.
[23] Mr Sarjeant says that in Montoux’s marketing materials, that company uses several of FIS’ Prophet variable names. Such variable names can be seen in an example demonstrating how its “Mode Copilot” works and in a video demonstrating Montoux’s “Excel Model Documentation”. He says that variables consist of computer code and a set of definitions. Allowing users to incorporate a variable’s computer code by simply referencing the variable’s name is said to save users from having to create the computer code relevant to that variable from scratch. Mr Sarjeant says the Prophet variable names were assigned by FIS using its own specific naming convention, which FIS recognises as watermarks or fingerprints. These demonstrate, he maintains, that Montoux obtained access to, and has used, FIS’s Confidential Information, specifically FIS’s confidential Prophet computer code.
[24] Mr Sarjeant contends that Montoux understands that the systems it copied or intended to copy are “proprietary”. This is because it describes them as “proprietary legacy platforms” and specifically refers to “proprietary actuarial languages like Prophet”. He notes that Montoux’s documentation states that its “Model Replication Module” works by ingesting input and output data without transforming it and then constructing a replica of an existing actuarial model, using artificial intelligence (AI). Mr Sarjeant acknowledges that the syntax of replicated code, according to Montoux, may not be identical to the syntax of the code ingested by the “Model Replication Module”. However, he argues the replicated model “provides the same output as the original when given the same input”.
Joseph Peterson’s evidence
[25] Joseph Peterson is Vice President of Actuarial Modelling at Lewis & Ellis Actuaries and Consultants, with 25 years of experience in the field of actuarial science
and the life insurance and annuity industry. His evidence was advanced by FIS in support of its claim here.
[26] Mr Peterson deposes that Prophet code, including many Prophet variable names and variables, are unique, original and distinct to Prophet. He explains that Prophet users build insurance products by incorporating desired variables from a library into their product. Each variable in Prophet has a unique variable name, with users utilising the variable names to incorporate the computer code associated with those variables into an insurance product. He says, because of certain limitations when Prophet was first developed (e.g. variable names limited to 12 characters, only letters, numbers and underscores, and all in uppercase), Prophet code for modelling was developed with several unique hallmarks to accommodate these limitations. He says some Prophet abbreviations are creative, unusual and not always intuitive, and even go against industry norms, with new users often requiring training on these abbreviations.
[27] Mr Peterson says the information he has reviewed has led him to conclude that Montoux has ingested Prophet model code and used it to create replica models. He understands references to “ingesting” input and output data, and to “source model” or “model documentation” to mean that Montoux feeds model code, like Prophet code, into its module replication module to copy those models. He says Montoux code uses many of the same abbreviation conventions as Prophet code, which are not intuitive or consistent with industry terminology. In his view they are instead consistent with Montoux’s ingestion (or copying) of Prophet code to replicate Prophet models.
[28] Mr Peterson says on 30 January 2024, he attended a Montoux product demonstration meeting which demonstrated how Montoux’s “Model CoPilot” ingested and analysed an example of a Prophet product’s code, and could answer questions about that code. He understood that Montoux was feeding Prophet code into its Model CoPilot tool and using it to generate outputs, some of which still contained the hallmarks of Prophet code.
[29] Mr Peterson’s overall view is that by ingesting/copying Prophet code and creating replica models, Montoux was able to bring its product to market more quickly
and with less development and testing than if it had truly been independently developed. This obviously meant that it avoided what were said to be thousands of hours, if not years, and millions of dollars in development costs.
[30] In his evidence in reply to the defendants’ evidence (some of which I canvass below), Mr Peterson deposes that it misses the point to say that use of Prophet “artefacts”, “model definitions”, “variable names” and “formulas” is not improper because Montoux did not directly obtain and copy Prophet source or object code. In this regard, he noted what were printouts of material from Prophet user’s code libraries—which included variables, actuarial definitions and formulas—in fact included thousands of lines of Prophet code. Mr Peterson also noted the copyright notice included in one of the first variables in Prophet’s actuarial code libraries. Accordingly, he suggested a licensed Prophet customer clearly would have understood the material to be copyrighted and confidential information of FIS.
Klaas Stijnen’s evidence
[31] Klaas Stijnen is a director of Montoux. He has a Master of Science degree from the University of Amsterdam, and it appears he has spent his entire professional life in the field of actuarial services.
[32] Mr Stijnen maintains that Montoux’s products were built on independent technologies. He asserts the Company never had any interest in copying Prophet as its plan was to produce a far superior product built on modern technology. He suggests here that FIS’s claims are unclear, but believes they are based on an assertion that a few variables names in Montoux’s publicly available information means Montoux copied or wrongly took Prophet’s systems for the development of its own. The variable names are not watermarks or fingerprints that demonstrate they stole Prophet computer code, according to Mr Stijnen. Nor are they ‘code’, as they are standard variable names used throughout the actuarial and financial services industries. He deposes there are hundreds, if not thousands, of publicly available documents utilising variables referred to in FIS’s claim. However, he says that FIS seems to think that once they bought Prophet in 2015, they could turn around and claim these variables were confidential and their property.
[33] Mr Stijnen contends that Montoux has never copied, nor had access to, Prophet source code or object code. Rather, Montoux has only ever received Prophet artefacts from Prophet customers looking to migrate away from Prophet. This has been in the form of PDF printouts of customers’ own model definitions which were solely used to support actuarial migration and development. He says these printouts did not contain source code or object code, but instead a series of variable names associated with an underlying actuarial formula. He notes clients had performed their own due diligence before sharing this information and were not induced by Montoux to do so as alleged. He explains that variables are not “code” but mathematical formulae which consist of largely standard actuarial equations. An observation he advanced was that the formulas as between Prophet and the Montoux platform were not perfectly identical. He says this is clear evidence that Montoux was not simply ingesting Prophet code and producing a copy of it. While the PDFs would allow Montoux to better understand the models that clients needed, he maintained Montoux would construct new models from scratch. Further, he contends the “model replication module” was just a mock-up for promotional reasons and was never operational.
[34] Mr Stijnen argues too that Mr Peterson is not independent. Instead, he claims Mr Peterson has a close relationship with FIS, having led the “enterprise actuarial modelling centre” at New York Life that partnered with FIS to build out a single actuarial platform. He also considers that Mr Peterson is not qualified to give an opinion on whether Montoux has copied Prophet code, as Mr Stijnen suggests he is not technically qualified or sufficiently astute to make such a judgement. He claims Mr Peterson misunderstands how the Model CoPilot operates, utilising as it does widely available Large Language Models without any fine tuning or training, and that it did not create derivative models trained on Prophet content.
William White’s evidence
[35] William White worked for Montoux for six years until its liquidation in November 2024. He held the role of Montoux’s Actuarial Solutions Architect at the time of its liquidation.
[36] Mr White expresses the view that Prophet was not copied by Montoux. He maintains they are two very different implementations using quite different technologies and approaches. Prophet contains libraries of variables and their mathematical definitions that are transcribed by the Prophet application into C code and compiled to an executable library which runs on a Windows operating system. In contrast, Montoux models are said to be expressed as a list of variable expressions akin to an Excel formula which are parsed and interpreted by a runtime engine written in Google’s Go running on Linux operating systems. As I understand it, Mr White says that Prophet’s variable definitions are run within an overarching processing loop that handles the time aspects of the actuarial projections. This differs from Montoux’s platform which treats each variable as an independent vector in memory and requires the model builder to ensure that the positional relationships between elements of each vector align as intended for the purpose of a financial projection model. Mr White contends these differences make it impossible for Montoux to ingest Prophet’s variable definitions, reformat them in Montoux, and execute them.
[37] Mr White argues that Mr Serjeant does not define “Prophet code” in his affidavit. He says if Mr Serjeant is referring to Prophet source code and object code, that would make no sense. This is because that would refer to the code required to build and compile the Prophet suite of applications that Prophet users utilise to manage and execute Prophet models. He says Montoux’s references to “Prophet Code Example” concern Prophet variable definitions which are expressions of mathematical equation used to calculate the value of each variable. He deposes that these represent the core actuarial mathematics and financial reporting regulations that actuaries constantly work with in their daily work. He says FIS has developed their collection of variable definitions representing standard actuarial formula from actuarial textbooks, published papers and regulatory documentation.
[38] In relation to material printed out and provided by Prophet customers to Montoux, Mr White says this simply amounted to customers providing Montoux with their best and most current documentation of the actuarial model they were using. While Prophet provides standard libraries and variables and their definitions, Mr White says users make their own significant amendments to standard variable definitions. This is done to implement the user’s own customisation to the models,
and to capture the bespoke features of the modelled insurance products. This means he suggests that in practice each Prophet customer only uses the standard library as a starting point for developing their models.
[39] Mr White also says the videos referred to in Mr Peterson’s affidavit are not evidence of Montoux copying Prophet. Rather, they show how a Prophet customer can identify variables that exist in their current Prophet Model, and they use Montoux’s AI tool to rationalise their model definitions to make them more efficient.
Legal principles
[40]The principles of interim injunctions are well-established:3
(a)there must be a serious question to be tried, with the plaintiff’s claim having at least “a tenable basis upon which it might be able to succeed at trial”;4
(b)the balance of convenience must favour the grant of the injunction, with the Court to assess the impact on the parties from either granting or refusing relief; and
(c)finally, an assessment of the overall justice of the position is required as a final check.
[41] FIS seeks an injunction on the basis there is a serious question to be tried in relation to breach of confidence and copyright infringement. For the Court to find a breach of confidence the following elements must be established:5
(a)the information has a necessary quality of confidence about it;
3 American Cyanamid Co v Ethicon Ltd [1975] AC 396, [1975] All ER 504 (HL); and Klissers Farmhouse Bakeries Ltd v Harvest Bakeries Ltd [1985] 2 NZLR 140 (CA).
4 NZ Tax Refunds Ltd v Brooks Homes Ltd [2013] NZCA 90, (2013) 13 TCLR 531 at [24] and [24].
5 Coco v AN Clark (Engineers) Ltd [1969] RPC 41 at 47–48 per Megarry J; Hosking v Runting
[2005] 1 NZLR 1 at [26]; and Hunt v A [2007] NZCA 332, [2008] 1 NZLR 368 at [65].
(b)the information has been imparted in circumstances importing an obligation of confidence; and
(c)there has been unauthorised use of that information to the detriment of the plaintiff.
[42]In relation to copyright infringement, a plaintiff must satisfy the Court that:6
(a)there is work in which copyright can subsist;
(b)copyright does subsist in the work, in that the work is a product of more than minimal skill and labour and is not a copy of another work;
(c)the plaintiff owns the copyright in the work; and
(d)copyright in the work has been infringed by:7
(i)the defendant copying all, or a qualitatively substantial part of the work;
(ii)there being sufficient objective similarity between the infringing work and the copyright work, or a substantial part of it; and
(iii)there being a causal connection, i.e. actual copying, or opportunity to copy.
[43] It is useful here to keep in mind the object of the present application for an interim injunction. This is outlined in McGechan on Procedure as follows:8
HR7.53.02 Object of an interlocutory injunction
The grant of an interlocutory injunction is a temporary and discretionary remedy. Its purpose is to protect plaintiffs against injury by violation of their
6 P S Johnson & Associates v Bucko Enterprises Ltd [1975] 1 NZLR 311 (SC) at 315.
7 Wham-O Manufacturing Co v Lincoln Industries Ltd [1984] 1 NZLR 641 (CA) at 666.
8 Jessica Gorman and others McGechan on Procedure (looseleaf ed, Thomson Reuters) at [HR7.53.02].
rights for which they could not be adequately compensated in damages recoverable in the action if the uncertainty were resolved in their favour at the trial. A plaintiff’s need for such protection must be weighed against a defendant’s need to be protected against injury resulting from being prevented from exercising legal rights for which the defendant could not be adequately compensated under the plaintiff’s undertaking in damages if the uncertainty were resolved in the defendant’s favour at the trial. The Court must weigh one need against another and determine where “the balance of convenience lies”: per Lord Diplock, American Cyanamid Co v Ethicon Ltd [1975] AC 396, [1975] 1 All ER 504 (HL) at 405–406, 508–509; Roseneath Holdings Ltd v
Grieve [2004] 2 NZLR 168 (CA) at [35].
[44] The application before me, being of an interim nature, is not one to determine the substantive merits of the plaintiffs’ claims. As McGechan on Procedure, in considering the purpose of an interim injunction, went on to say at HR7.53.02:
In Commerce Commission v Viagogo AG [2019] NZCA 472, [2019] 3 NZLR 559 at [31], the Court of Appeal described the purpose of an interim injunction as to improve the chance of the court being able to do justice after a determination of the merits at the trial. Citing Lord Hoffmann in National Commercial Bank Jamaica Ltd v Olint Corp Ltd [2009] UKPC 16, [2009] 1 WLR 1405 at [16] and [17], the Court said the basic principle is for a court to take whichever course seems likely to cause the least irremediable prejudice to one party or another.
Is there a serious question to be tried?
Plaintiffs’ submissions
[45] Mr Murray for FIS submits that the evidence, even at this early stage, gives rise to a strongly arguable case that Montoux has improperly accessed, used, copied and adapted substantial parts of FIS materials. These include the Copyright Works and Confidential Information. He says that, consequently, the software and code held by Montoux contains work that belongs to FIS as well as FIS trade secrets. He submits that:
(a)the defendants do not deny that Montoux has accessed, used, copied and/or adopted FIS’s material, with their statements that the code underlying certain “products” followed an independent design path being unsupported by any particulars or evidence; and
(b)Montoux fails to explain why its publicly available documentation refer to “replication” and reference use and “ingestion” of Prophet, only
saying that Montoux was “not simply ingesting Prophet Code” and that formulae used by Montoux were not “perfectly identical” to Prophet.
[46] Mr Murray submits that Montoux’s evidence first, that FIS clients who disclosed materials to Montoux “performed their own due diligence” and secondly, that other persons allegedly also had access to Prophet documentation, does not provide a defence to breach of confidence or copyright. Montoux’s further reasoning that the information shared with Montoux was not copyrightable or confidential to FIS is said by Mr Murray to be a flawed opinion simply unsupported by the facts.
[47] Mr Murray goes on to maintain too that FIS’s Confidential Information has a necessary quality of confidence about it, given the measures taken to secure it and the limits placed on its access and use. He contends that the defendants were aware that the Prophet artefacts Montoux received from FIS customers were proprietary and confidential to FIS. He notes that FIS customers were expressly forbidden from sharing “outputs” with any FIS competitor who was not authorised by FIS to receive that material. Those outputs included documents, reports, statements and other outputs from the customer’s own work constructing actuarial models. He claims these customers were induced by Montoux to breach their contracts with FIS. He argues too that both defendants received the Confidential Information in circumstances where they knew or ought to have known it was divulged in breach of an obligation of confidence owed to FIS.
[48] FIS’s position here is that there has also been copyright infringement, as copyright programmes, which reflect formulae, variables, algorithms and logic structures, are recognised as literary works capable of attracting copyright. Mr Murray notes FIS and its predecessors have invested significant skill and labour into developing Prophet. This product, it is claimed, is highly original. FIS has long owned Prophet, and it submits that Montoux’s public materials, and recent admission that it uses Prophet artefacts in its software, must indicate that Montoux has copied a qualitatively substantial part of Prophet. Mr Murray says too that Montoux’s business of replicating software, asking customers to provide Prophet artefacts, migrating customers away from Prophet by copying Prophet code, and hiring Prophet specialists, clearly demonstrate a causal connection here.
[49] Furthermore, Mr Murray submits that Montoux’s request for particulars is simply an attempt to confuse matters. It does so by asserting the statement of claim is unclear, by requesting additional evidence already in its possession, and by demanding additional details not relevant to FIS’s claim. He notes that in Medtech Ltd v Valentia Technologies (NZ) Ltd,9 the Court found that although particularisation is of critical importance in intellectual property cases, an overly forensic approach to the need for particulars should not be taken. Rather, claims are to be addressed on the evidence filed.
Defendants’ submissions
[50] In response, Mr Snape, for the defendants, appears to rely significantly on the decisions of the High Court and Court of Appeal in Fisher & Paykel Financial Services Ltd v Karum Group LLC.10 In that case, Karum alleged that Fisher & Paykel used its licence access to copy Karum’s CMS, and as a result breached its confidentiality and copyright. Its claim was that Fisher & Paykel copied non-literal elements of its CMS, such as the logic inherent in certain components, to replicate the processing logic. In the High Court decision, Rodney Hansen J noted that, in reference to the TRIPS Agreement to which New Zealand is a party, copyright protection will not extend to ideas, procedures, methods of operation or mathematical concepts as such.11 The Court of Appeal went on to note that the literal elements of a computer program protectable by copyright normally comprise the source and object code.12 The functionality of a computer program may be replicated precisely by a completely different program, and cannot be protected by copyright.13 The Court noted the findings in Nova Products Ltd that a written work consisting of the specifications of the functions of a computer program would attract protection as a literary work, but the functions themselves would not.14 Reference was also made to SAS Institute, where it was found that studying technical manuals describing operation and use of
9 Medtech Ltd v Valentia Technologies (NZ) Ltd [2024] NZHC 2107 at [74](e).
10 Fisher & Paykel Financial Services Ltd v Karum Group LLC [2012] NZHC 3314, [2013] 2 NZLR 266 [Karum HC judgment]; and Karum Group LLC v Fisher & Paykel Financial Services Ltd [2014] NZCA 389, [2014] 3 NZLR 421 [Karum CA judgment].
11 Karum HC judgment at [177].
12 Karum CA judgment at [101].
13 At [108] citing Navitaire Inc v easyJet Airline Co Ltd [2004] EWHC 1725, [2006] RPC 3 at [125]; and Nova Productions Ltd v Mazooma Games Ltd [2007] EWCA Civ 219, [2007] Bus LR 1032 at [55].
14 Karum CA judgment at [162] citing Nova Productions Ltd v Mazooma Games Ltd, above n 13.
various components, and replicating a large part of the functionality of the components, does not infringe copyright.15
[51] Mr Snape argues that, while copyright protection for non-literal elements in computer programmes is technically possible, in the absence of source code and/or object code being copied, the practical reality is that a copyright claim is doomed to fail.
[52] He further submits that the narrow ambit for claiming copyright protection over computer programmes ties in with the rationale for proper particularisation in intellectual property cases. He refers to the findings in Ocular Sciences Ltd v Aspect Vision Care Ltd,16 in which the Court found that if a plaintiff wishes to seek relief against a defendant for misuse of confidential information, it is the plaintiff’s duty to ensure the defendant knows what information is in issue. The Court in that case also noted that unless the confidential information is properly identified, an injunction in such terms is of uncertain scope and may be difficult to enforce.
[53] The complaint is made that FIS here fails to explain and describe the constituent parts of Prophet, and that their broad inferences do not match up with the definitions in Karum. Mr Snape notes Mr Serjeant’s references to “Prophet’s computer code”, which he says includes formulae, structures and logic which are non- literal, and are not copyright protected. He observes that Prophet’s source code and object code are held by FIS safely under lock and key. In his submission, whatever Prophet’s customers have provided to Montoux, it simply is not source code or object code. Mr Snape also submits that variable names, although referred to by Mr Serjeant in his second affidavit, are not capable of copyright protection. He argues that if Montoux has not copied the literal elements of Prophet, namely the source/object code, the plaintiffs have no viable copyright claim. It follows that there is no basis to assert Montoux was not perfectly entitled to have reference to the functional and structural elements from Prophet.
15 At [115] and [163] citing SAS Institute v The World Programming Ltd [2010] EWHC 1829, [2011] RPC 1.
16 Ocular Sciences Ltd v Aspects Vision Care Ltd [1997] RPC 289 (Ch).
[54] With regard to the claim for breach of confidence, Mr Snape says Mr Stijnen’s evidence explains how the allegedly confidential information is available in the public domain, with the underlying variable definitions being core actuarial mathematics in public use. He argues FIS’s evidence in reply makes no attempt to deal with or explain how Mr Stijnen had found Prophet’s confidential information simply by internet search. He further contends that FIS fails to address the fact that Prophet users customise their own libraries. Client amendments and data also, according to Mr Snape, will not be the plaintiffs’ confidential information.
Analysis
[55] At the outset, I need to say that the very technical nature of the evidence on which FIS’s claims and Montoux’s defence are based makes even a broad preliminary assessment of the possible merits of the substantive application somewhat difficult. I note also that the discovery process here is said to be incomplete. There are unresolved claims also around a lack of particulars in FIS’s pleadings. With that context in mind, I consider that, at this early stage, there is a serious question to be tried, especially with regard to the breach of confidence issues. I now set out my reasons for this conclusion.
[56] As to the breach of copyright cause of action, I accept Mr Snape’s submissions that to some degree inconsistent terminology has been used by FIS in regard to its claim over what information or works Montoux has actually accessed and used from FIS and Prophet. I accept that FIS has referred to “Prophet Code” and “Prophet Computer Code”, and has noted that such code includes source code, object code, formulae, structures, and logic. However, in my view, FIS has confused the issue to an extent in not properly clarifying what parts of the Prophet Code Montoux has accessed and used in its own software, and more importantly, what parts it has not. This may be relevant given the distinction made in Karum between literal and non-literal elements of a computer programme, with copyright infringement in respect of non-literal elements much more difficult to demonstrate. As noted in Karum, the literal elements are generally taken to be the source and object code,17 and it is uncontroversial that such elements are subject to copyright protection.18
17 Karum CA judgment, above n 10, at [101].
18 International Business Machines Corp v Computer Imports Ltd [1989] 2 NZLR 395 (HC) at 411– 413.
[57] However, on one argument here, FIS may not be alleging that Montoux has accessed and copied Prophet’s source and object code. Rather, it could be suggested that FIS’s claim is only something of a lesser nature, being that Montoux is using variables, actuarial definitions, and formulae from Prophet, which have been described as “Prophet artefacts”. And, there perhaps remains a question as to whether variables, actuarial definitions and formulae are non-literal or literal elements. In Karum, non-literal elements are said to include the program’s structure, namely the programmer’s choice of subroutines, relationships and prescribed interactions between them.19 This does not appear to align with Mr Sarjeant’s definition of variables, which he describes as computer code and a set of definitions. Nor does Mr Stijnen’s description, namely that they are mathematical formulae consisting of actuarial equations, seem to fit into the non-literal category. Mr Murray’s written submissions state that source code consists, in part, of thousands of variable names and actuarial formulae. It could therefore be inferred that the Prophet artefacts received by Montoux from Prophet customers did in fact contain at least part of the Prophet source code. Accordingly, it must follow that it remains reasonably arguable that these variables, actuarial definitions and formulae are in fact literal elements.
[58] However, this does not address the evidence Montoux has purported to advance that the variables are not distinct to FIS or Prophet and are in fact widely used throughout the actuarial and financial services industries. It also fails to respond to the contention that FIS’s collection of variable definitions are derived from publicly available sources such as actuarial textbooks, published papers and regulatory documentation. Nor does it address the evidence that users make significant amendments to FIS’s standard variable definitions, meaning these only act as a starting point for customers to develop their own models. This could perhaps cause some issues here in establishing parts of the test for copyright infringement. In particular, that the Copyright Work is a copy of other work, or that there is sufficient objective similarity between the copyright work and the infringing work, or a substantial part of it. Additionally, in my view, it may well be problematic to say that the use of Prophet naming conventions in variable names used by Montoux is necessarily sufficient to demonstrate that Montoux is using FIS’s copyright protected work.
19 Karum CA judgment, above n 10, at [102].
[59] Regardless, an interim injunction is clearly available to restrain an infringement of copyright. It is often a useful remedy for protecting a plaintiff’s intellectual property interests where, like here, damages would often be inadequate. At this early stage, it is arguable there has been an infringement of FIS’s copyright in its Prophet software, although I do not need to express a firm view on that here. I leave on one side that breach of copyright claim. It will remain for fuller consideration when the permanent injunction hearing between the parties takes place.
[60] I turn now to FIS’s breach of confidence cause of action. On that aspect, in my view, there is an arguable claim which raises a serious question to be tried in this case. It is true that, as with the claims of copyright infringement, there are questions around to what degree the variables, definitions and formulae are different from that in the public domain and widely used in the industry. However, it is clear that FIS has substantial measures in place to protect its Prophet software. It is relevant, as I see it, that FIS license agreements prohibit FIS customers from sharing outputs (including from the customer’s own work constructing actuarial models), with competitors or third parties in general. In my view, this establishes at least a tenable basis that the variables, definitions and formulae were imparted from FIS to its customers in circumstances importing an obligation of confidence. It follows that, by receiving this information/work from FIS customers, Montoux, as a competitor in the industry, was or should have been aware that this was confidential information. Montoux then arguably went on to use that information to FIS’s detriment, namely towards its own rival actuary software. I do not consider that the fact FIS discovered the alleged confidential information through an online search means the information is not confidential. Rather, it acts as a potential example of a breach of confidence, in that Montoux was uploading onto a public website videos explaining its software which used what FIS says is its confidential information.
[61] Accordingly, I am satisfied here that there is a serious question to be tried in this case.
Does the balance of convenience and overall justice favour granting the injunction?
Plaintiffs’ submissions
[62] On this aspect, which involves carefully weighing the respective positions of the parties, Mr Murray submits that FIS will suffer irreparable harm if the interim injunction is not granted. If Montoux and/or the Liquidator were to promote, sell or dispose of the relevant software products, the damage to FIS he says could be catastrophic, with its entire investment in Prophet being put at risk. He considers too that damages here would be nearly impossible to quantify, not to mention the difficulty/impossibility in recovering these from a company in liquidation. If the interim injunction is not granted, and the Liquidator was to sell the software assets, he says FIS’s application and much of its substantive proceeding would also be rendered nugatory. FIS would have to consider a possible new proceeding and an interim injunction application against the purchaser.
[63] Mr Murray goes on to say that there is also no evidence here that an interim injunction will destroy the value of any Montoux assets, or render Montoux’s intellectual property valueless. He says there is simply no evidence before the Court as to the value of the Montoux software and how or why it will be eroded over time. The value of the relevant software assets is already compromised, Mr Murray says, given the argument that they are unlawful copies of FIS’s Prophet software. It is noted too that the defendants voluntarily undertook not to advertise, publish or sell the Copyright Works and the Confidential Information. He also points out the Liquidator offered to extend the undertaking until unspecified dates. Mr Murray says the undertaking had been in place for 74 days as at 29 July 2025, with no evidence of harm to Montoux. If the interim injunction was to be reversed later, he says any damage to Montoux could be addressed through FIS’s undertaking as to damages. Mr Murray also notes an interim injunction will simply maintain the status quo pending determination of the substantive application here.
[64] The overall justice is also said to favour the granting of interim orders here. Mr Murray says there has been no delay by FIS. He notes that mere months after discovering the evidence of misuse of Prophet code, FIS initiated proceedings in
Delaware against Montoux. He argues that prior to its liquidation, FIS was not aware of any potential sale of Montoux software assets, but once the Liquidator was appointed, a sale became likely. FIS, he maintains, then responsibly engaged in open discussions for over five months. The need for an interim injunction to restrain a sale was only triggered when the Liquidator threatened to dispose of the relevant assets after 1 May 2025. This meant that FIS had no option but to initiate proceedings, which it did on 6 May 2025.
[65] Mr Murray further submits that while FIS has acted reasonably throughout, the Liquidator adopted Montoux’s position wholesale, dismissed FIS’s claims as baseless and vexatious, and threatened to release FIS’s confidential and copyright information via sale. The expert report relied on by the Liquidator was commissioned weeks after the Liquidator’s immediate characterisation of FIS’s claims as baseless and vexatious. Mr Murray also maintains that the Liquidator has refused to engage with FIS’s proposals to avoid litigation, and has not put forward any counterproposal that would allow the issues to be resolved. He notes liquidators are required to be independent and be seen to be independent and cannot advocate for or advance the interests of one group of stakeholders in a dispute against another. Instead, in this case, Mr Murray suggests that the Liquidator has adopted the shareholders’ positions and opinions, arguably relying on the input of Mr Stijnen. He adds all this should help to support the position that the overall justice here rests with FIS’s application succeeding.
Defendants’ submissions
[66] In his submissions, Mr Snape refers to the following commentary from Sim’s Court Practice:20
The material before the court at the time of hearing an application for an interlocutory injunction is untested. The decision to grant an interlocutory injunction which effectively overlaps with and would pre-empt the ultimate relief sought, should be considered only with the utmost caution. Such an order should be made only when the balance of advantage plainly favours the grant of relief: Channel Tunnel Group Ltd v Balfour Beatty Construction Ltd [1993] 1 All ER 664, 690, per Lord Mustill (HL).
20 Laura O’Gorman (ed) Sim’s Court Practice (online ed, LexisNexis) at [HCR7.53.10(h)].
[67] Mr Snape submits that there has been significant delay in this case. This is so, he says, given FIS acknowledges it allegedly discovered and investigated an example of “Prophet Code” being copied in June 2024, but did not bring its application for interim injunctive relief until 6 May 2025. He notes also that FIS waited until 1 November 2024 to file the Delaware complaint, with no cease and desist letters sent in the meantime. He observes that the status quo position whereby Montoux has been unable to deal with its actuarial software has been in place for a year now. In his submission, the plaintiffs have not addressed any reasons for their delay in seeking interim relief.
[68] Mr Snape also suggests that there is no real evidence here to support the claim of irreparable harm to FIS if the interim injunction is not granted. He argues that if there was a true need for protection, FIS would not have sat around for 11 months before seeking injunctive relief. No attempt is made to address the practical reality that because of that delay, if Montoux had in fact had confidential or copyright material, “the cat” would already be “well out of the bag”. FIS, he says, is a massive, diversified multi-national company, and nothing has come close to showing to any degree that Montoux’s software will impact FIS’s operations. He argues no damage has been suffered to date, with the best that can be said being that if Montoux sold its software there would be a competing actuarial software programme in the market. In contrast, Mr Snape contends there will be irreparable harm to Montoux and its creditors if the injunction is granted. This is because, in his words, it will “massively” reduce the value of its asset, if not render its software valueless.
[69] Lastly, Mr Snape says FIS’s conduct here has been “appalling”, given they filed proceedings against the plaintiffs in Delaware without warning months after viewing the information publicly uploaded by Montoux. Then, having belatedly realised the Liquidator had a duty to realise Montoux’s assets, they brought the present claims and injunction application. He also claims FIS has repeatedly shifted the goalposts of what their claims actually are, whilst vehemently opposing giving further particulars. He takes particular exception to the claims that the Liquidator failed to act in an impartial manner. In his submission, there is simply no proper foundation advanced in the pleadings or evidence for the Liquidator doing so. All this conduct, Mr Snape
complains, represents “harassment and oppression” against the defendants “of the worst kind”.
Analysis
[70] At the outset, I need to say that I treat with some caution Mr Snape’s claims outlined at [68] and [69] above. And, I agree with Mr Murray that the balance of convenience here clearly favours FIS. A failure to grant the interim injunction would enable the Liquidator to progress with the liquidation and sell assets said to be Montoux’s, which would render moot the substantive application here for a permanent injunction. If FIS in fact has a valid claim and Montoux’s software does contain its copyright-protected or confidential works/material, then FIS would certainly face loss.
[71] Maintaining the injunction on an interim basis would also mean largely maintaining the status quo pending final determination of the injunction application. I note here, however, that the orders sought are more expansive than the undertaking given before La Hood J, as it would prevent the Liquidator from dealing with the alleged Copyright Works and Confidential Information in any manner, rather than just preventing advertisement, publication or sale.
[72] I accept that Montoux, and by extension its creditors, may be affected by the Liquidator being unable to deal with its software and progress the liquidation. While it was not specifically raised with me by counsel in the hearing, but rather after the fact, Montoux “remains trading” in the sense that it continues to service a single contract with a customer. The broader orders sought here would mean those services may no longer be able to be fully provided. It is said this would remove an income stream that the Liquidator claims benefits the liquidation and its ongoing costs. However, the defendants have declined to provide details of what services are being provided, what income Montoux is deriving from this trading, and how the loss of this income will impact the liquidation. Accordingly, while granting the injunction may prejudice Montoux to an extent, the degree of prejudice with respect to this trading is unclear. Furthermore, when there is a tenable basis to say Montoux’s assets contain copyright or confidential material, it is difficult to accept the argument that Montoux
faces significant prejudice in the short term in being prevented from trading in those assets.
[73] Additionally, as I see the position, there does not appear to be any actual evidence at this stage that continuing the status quo for a short time will mean that Montoux’s assets will lose significant value or that they will be rendered valueless. Continuing the injunction will simply mean the software assets will remain with the Liquidator unsold in the meantime. I also consider it relevant that Montoux has been willing in the past to give undertakings not to advertise, publish or sell its material that is alleged to contain copyright protected or confidential information. It has also previously sought an adjournment delaying, to some extent, the matters here. If continuing the status quo was really causing Montoux considerable losses, one would expect both the Company and the Liquidator to oppose any delay in determining the permanent injunction application.
[74] There has also not been significant delay on the part of FIS in this case. The application for injunctive relief was brought on 6 May 2025, and not earlier, because it appears it was not until 25 March 2025 that the Liquidator indicated he intended to sell Montoux’s assets. As identified by Mr Snape, it is true that there was a delay of a number of months between the discovery of the alleged examples of copyright infringement and breach of confidentiality and the filing in Delaware of the claim against Montoux. However, I consider that the delay can be explained by FIS needing time to examine the video in question, and to determine whether or not it should bring a claim. Between November 2024 and May 2025, there were also some discussions between the parties and attempts made by FIS to seek to resolve the matter out of court.
[75] I do not consider the circumstances here to be similar to the case relied on by the defendants, namely On-Line Digital Solutions Ltd v Riddick.21 That case concerned breach of a restraint of trade, where the plaintiff had waited nine-months, allowing the defendant to establish and operate a business, before seeking an injunction.22 In contrast, FIS took action on a much less clear potential breach of
21 On-Line Digital Solutions Ltd v Riddick [2021] NZHC 3199.
22 At [112].
copyright and/or confidence within four months, and was engaging with Montoux during the period between bringing the US proceeding and bringing this proceeding. Accordingly, it cannot be said that there was unreasonable delay here justifying a refusal of injunctive relief on an interim basis.
[76] I also do not consider that FIS’s conduct has been “appalling” and “of the worst kind” as claimed by the defendants. I accept that it is undesirable for claims of impartiality to be made against a licensed insolvency practitioner such as the Liquidator. It would have also been preferable here for FIS to have served Montoux with a cease and desist letter before bringing the proceedings against it. However, it is unlikely such a letter would have meant these proceedings would have been avoided, when it does seem to me that the defendants have been generally opposed to taking steps to resolve matters outside of court. In my view, overall, the general conduct of the parties is a largely neutral factor in relation to where the balance of convenience and overall justice lies. I certainly do not share the view that FIS’s conduct amounts to harassment and oppression of the worst kind, or has any similarities to the weaponisation of a vague breach of confidence claim described in Ocular Sciences Ltd.23
[77] By way of an aside at this point, I interpolate what Mr Murray proposed in his concluding remarks that what FIS seeks overall is that the status quo should remain to give an opportunity for independent experts by agreement to look at the material and to determine the extent of FIS’s copyright in it. In addition, he said the option might arise that certain aspects could be extracted from this material that clearly infringed the copyright. He proposed that a sale could then be allowed of the remaining material which is not subject to copyright. This would provide more certainty for all parties concerned, and it seems to me to be an idea worth exploring.
[78] Finally, I acknowledge the commentary that Mr Snape has helpfully referred to that where there is a substantive application for a permanent injunction, an interim injunction should only be made where the balance of advantage plainly favours the grant of relief. However, in this case, I consider that threshold has been met.
23 Ocular Sciences Ltd, above n 16, at 360.
Conclusion
[79] For all the reasons I have outlined above, the application by FIS for an interim injunction succeeds. Orders will follow.
[80] Although the circumstances here do not meet the threshold for allocating a priority fixture,24 I comment that it is appropriate for the substantive injunction application to be set down for an early hearing. This will enable matters to be brought to a head and for the liquidation of Montoux to progress.
Result
[81] Given FIS’s application for an interim injunction has succeeded, the following orders are now made:
(a)Leave to commence this proceeding is granted.
(b)Until further order of this Court, an interim injunction is granted preventing the defendants from publishing, copying, communicating, promoting, adapting, using, selling, realising, disposing of, or otherwise dealing with any works that are, or are alleged by the plaintiffs to be, direct or indirect copies of, or that otherwise copy, adapt, incorporate, or use any part of the plaintiffs’ Copyright Works or Confidential Information (as defined in FIS’s application), including:
(i)the software products, including actuarial modelling software and computer code, held or controlled by the defendants, including:
1.the software product or products advertised by the first defendant as the “Decision Science and Actuarial Automation Platform”, “the Platform”, “the Montoux Actuarial Platform”, and/or “Montoux’s actuarial software”;
24 See Gavin v Powell [2020] NZHC 1224 at [25].
2.the artificial intelligence products and replication modules, including the software product advertised by the first defendant as the “Model Replication Module” or the “AI Driven Replication Toolkit”;
3.all software products and other works advertised by the first defendant in the YouTube videos titled “Model Copilot Excel Demo”, “Montoux Github Copilot” and “Montoux Valuation Workflow Demo”;
4.software products that have ingested and/or are “trained” on all or part of the plaintiffs’ Copyright Works or Confidential Information;
5.all other Infringing Copies (as defined in FIS’s application),
(ii)manuals, instructions, guidelines, and other works that relate to such software products;
(iii)advertising and promotional materials that relate to such software products, including “blog” posts and website articles.
Costs
[82] Costs are reserved at this point. It is suggested they might be deferred and determined as part of the substantive injunction hearing to follow.
[83] Alternatively, if the parties require a costs determination now and are unable to agree this issue between themselves, then they may file memoranda (maximum four
pages each) and the Court will determine the issue of costs here on the papers.
Gendall J
Solicitors:
A J Park, Wellington for Plaintiffs
Thomas Dewar Sziranyi Lett, Lower Hutt for Defendants
1
7
0