Q&A: Technology disputes
June 2025 | SPECIAL REPORT: INTERNATIONAL DISPUTE RESOLUTION
Financier Worldwide Magazine
June 2025 Issue
FW discusses technology disputes with Rachel Glass, Jeremy Sharman, Matthew Pack and Anne-Sophie Lampe at Bird & Bird LLP.
FW: What types of technology disputes are you currently seeing?
Glass: We are continuing to see a large number of technology project-related disputes. Frequently, these relate to issues which have arisen while the project remains underway, meaning that the challenge is to support the client – either on the customer or supplier side – with getting the project back on track while preserving their rights or settling claims. The second main type of dispute we see is licensing-related, where allegations of over-use are made by a software supplier, often following an audit process. Again, the driver for both parties is normally to resolve matters commercially so that the relationship can continue. Where the parties have already parted company, the dynamic can be less constructive and the licensing dispute more difficult to resolve without resorting to formal dispute resolution processes.
Sharman: We see technology disputes impacting projects concerned with critical national infrastructure. These include communication networks, electronic payment systems and the delivery of new energy infrastructure. Inevitably, because of their sensitivity, these tend to be resolved on a confidential basis. There was also a period when the transition of services to the cloud generated disputes – for example point of sale and booking systems being moved to the cloud in the hospitality sector – but we are seeing less of that now.
Pack: Projects integrating multiple legacy IT systems, platforms and processes are an area ripe for disputes. This is particularly so in banking and financial services, which is both highly regulated – including the resilience of providers’ IT systems – and where we are seeing increasing vertical consolidation, particularly in the FinTech space. Even where integration itself goes smoothly, multiple legacy systems operating in parallel also give rise to in-life challenges. This can also increase costs and the number of disputes longer term, both in terms of systems failures and adapting multiple systems to changing regulation and external threats, particularly around cyber security and resilience.
Lampe: We see an increasing number of pre-litigation issues concerning artificial intelligence (AI) implementation projects where the AI system provided does not meet the client’s expectations. We have noticed that often the cause of the difficulties relates to the lack of quality of the training data. The issue is then to determine which party should be liable for the quality of the training data which is required for the AI system. Providers tend to consider that the quality of the training data is the client’s responsibility, while the client considers the provider should have better advised them about this.
“We are witnessing a rise in regulatory-driven conflicts, especially following investigations by regulators into the practices of tech companies. Additionally, there is more and more litigation between tech companies themselves based on competition law rules.”
FW: Could you provide an overview of notable trends heightening the risk of technology disputes?
Glass: Software licensing is impacted by the continuing evolution of licensing metrics, meaning that the rights enshrined in longstanding agreements – which may have been ignored by the licensor and licensee for some time – may bear little resemblance to the virtualised estate of the customer many years later. In those circumstances, allegations of over-use become a matter on which technical expertise is required as well as detailed knowledge of the licensed estate. The general trend among suppliers toward subscription-based licensing is also a particular challenge for customers, who may find their supplier unwilling to continue to renew maintenance arrangements unless a subscription approach is adopted, regardless of the terms of the original agreement.
Lampe: As in the UK, we have seen an increase in disputes arising out of software licensing agreements in France. In particular, in a recent case where the software publisher had unilaterally changed – through a press release – the commercial terms of their licences which enabled the customer to order additional perpetual licences to a subscription model, the court ordered the software publisher to perform their licence agreement and required them to grant their customer new perpetual licences. This is on the basis that software publishers are not entitled to unilaterally change the commercial terms of their licence to a subscription model.
Sharman: It is clear that the move to a digitised world has led to issues with legacy systems particularly where hybrid solutions are proposed without sufficient planning or testing. In addition, the advent of AI and its reliance on different data sets, whether provided by a third party, the client or its counterparty, increases contractual complexity and the likelihood of disputes when things go wrong. AI and other ‘black box’ technology, which is increasingly used to determine outcomes, for example in the evaluation of procurement bids, is also likely to give rise to disputes, particularly where it is not possible to understand fully how decisions were reached.
Pack: AI’s day to day impact has been huge in the public consciousness, but still relatively concentrated in the well-known AI large language models, particularly foundation models. This has limited the number of ‘pure’ AI disputes that have emerged so far or, more specifically, those that are being dealt with in public forums. As AI models and systems proliferate and more developers take their offering to market or the public, the prospect of AI-linked disputes will inevitably increase.
“Clauses in contracts which say that inaction is no waiver of a right or claim are not an absolute protection – doing nothing or failing to take action at the right time may lead to a loss of rights entirely.”
FW: How have evolving rules and regulations led to increased scrutiny of technical innovation? To what extent is this dynamic leading to a rise in disputes?
Pack: The innovation of all things AI-related, for which regulators are developing policies and standards at different speeds, has led to a huge increase in the scrutiny of technical innovation. I say regulators in the broadest sense, because they are not only governments: influential trade and standards bodies, and international organisations such as the European Union (EU) and the Organisation for Economic Co-operation and Development all recognise the monumental importance of AI standards and adapting existing or building new legal frameworks to address them. A key issue going forward will be the extent to which, in private contractual arrangements, liability for damage caused by AI systems and their decisions or analysis will be regulated, how consistent this will be, or whether it will fall to parties to allocate the risk between themselves. This process will inevitably lead to more scrutiny of how developments in AI are understood, how they can cause harm, and who or what should be liable in respect of that harm.
Lampe: The increasing regulation of the tech sector, particularly in areas like AI, data privacy, cyber security and digital finance, is leading to new forms of disputes for the industry. These are adding to traditional IT contractual disputes. Notably, we are witnessing a rise in regulatory-driven conflicts, especially following investigations by regulators into the practices of tech companies. Additionally, there is more and more litigation between tech companies themselves based on competition law rules. It has also become common in IT contracts to emphasise providers’ regulatory compliance obligations, leading clients to invoke regulatory breaches to evidence contractual faults against them. The intensification of regulation in the tech sector also enables consumers to initiate legal actions for non-compliance with regulations. This is particularly evident in collective actions, now very common in Germany, relating to General Data Protection Regulation violations. This trend is especially relevant within the EU, notably with the recent amendment to the defective product directive, which now includes software within its scope, and the ongoing discussions around potential AI liability regulations. In addition, it is likely that some provisions of the European AI Act would invite parties with a legitimate interest, including copyright holders, to exercise and enforce their rights and take direct action against generative AI model providers who they allege fail to meet their transparency obligations under the EU AI Act. Furthermore, the complexities in interpreting various regulations and the questions of applicable laws and competent jurisdictions in cross-border contexts further complicate the situation and increase the risk of disputes.
FW: Could you highlight recent technology disputes that grabbed your attention? What lessons can be learned from their resolution?
Glass: The dispute between Tata Consultancy Services and the Disclosure Barring Service (DBS) in the England & Wales Technology and Construction Court was interesting for a number of reasons. First, it was a case which looked closely at the proper English-law interpretation of a limitation and exclusion of liability clause, which is a frequent battleground in IT project cases. Second, it was a dispute in which – like many failed IT projects – both sides felt aggrieved: Tata brought a claim for £110m, while DBS counterclaimed for £109m. After almost a month in trial, in a 100-page judgment the court ultimately found that each party was successful on different aspects of the dispute, meaning that the net amount payable by DBS to Tata was only £5m. It is therefore an illustration of the highly complex nature of many technology project disputes, and the potential value of commercial settlement over pursuing such cases to trial.
Sharman: An important issue, and one that was recently highlighted in a case involving IBM and Switzerland-based software company LzLabs GmbH, concerns reverse engineering of software. Software vendors are naturally protective of their products and keen to ensure that licensees do not use those products unlawfully by using invasive reverse-engineering techniques to create something for their own benefit. This case related to LzLabs’ development of the Software Defined Mainframe (SDM) platform, which enabled customers to transfer existing mainframe software to modern computer architectures such a Linux, potentially reducing reliance on IBM’s proprietary systems. IBM claimed that LzLabs’ subsidiary Winsopia had unlawfully reverse engineered its mainframe software to produce the SDM after purchasing an IBM mainframe computer in 2013 and entering into a licence agreement with IBM. LzLabs argued that only “observation, studying and testing” was undertaken in compliance with the EU Software Directive. However, the court found in favour of IBM, ruling that Winsopia had breached the terms of the IBM software licence. The case contains an important reminder to software vendors to ensure that their licences contain clear limitations on the use of the software, to prevent reverse engineering, and so avoid any arguments as to whether any software reconfiguration is permitted.
“It is clear that the move to a digitised world has led to issues with legacy systems particularly where hybrid solutions are proposed without sufficient planning or testing.”
FW: Once a technology dispute has surfaced, what options should be considered to resolve it? How important is the early engagement of experts?
Glass: If a dispute has arisen during the parties’ ongoing relationship – for example in relation to an IT project, during the implementation phase or an ongoing maintenance and support phase – parties will often be looking for rapid resolution to enable them to move on with the project or relationship. There are a number of ways in which this can be achieved. One is commercial negotiations, potentially through a series of escalations to more senior management or through a more structured mediation exercise, ideally with a mediator with some experience of these types of disputes. Expert determination may be suitable for disputes of a discrete technical or financial nature. Resolution by a ‘dispute board’ or adjudication-type procedure is another option. Ultimately, resolution by a court, or by consent of the parties, by arbitration, is also available.
Pack: In some industries and sectors, it is common to find ‘fix first, argue later’ terms and procedures which are agreed to enable a rapid decision on liability so that a project can quickly get back on track. Construction is a good example, where in the UK there has been a statutory adjudication scheme for decades, and parties to M&A transactions will frequently agree an expert determination process to deal with post-completion disputes such as deferred consideration and warranty claims. This is far less common in technology disputes, and needs to be considered at the time of contracting rather than as disputes arise, to achieve the goal of a rapid answer so that parties know where they stand and can progress the project.
Sharman: The early involvement of an expert can be important depending on the nature of the dispute. Where the client is the technology supplier they will inevitably have considerable expertise when it comes to their solution, which may mean, in the early stages at least when the focus is on resolution or the preparation of a claim, the separate engagement of an independent expert is not necessary. That said, if the case progresses, having an independent expert to validate the client’s position, even if the client is the supplier, will be important. Early engagement is also advisable in cases where there may be a limited number of individuals with the requisite expertise. As a related point, having experts involved at an early stage who can agree a factual description of the solution architecture may help reduce the number of disputed issues and the extent of any disclosure.
“A lot can be gained by using contractual mechanics, such as relief notices and change procedures, to protect a company’s position.”
FW: What advice would you offer companies on steps they can take to reduce potential technology disputes, including during contract negotiations?
Sharman: So many disputes arise because there is inadequate planning prior to contract signature. The pressure to sign a contract, including an overly optimistic view on the utility of high-level specifications and an unwillingness to deal with difficult issues, may mean that at the time of contract signature there is insufficient certainty around key deliverables and their associated specifications. This can lead to problems, including delay and additional costs, when the project comes to the testing and implementation phases. So a key piece of advice would be to spend adequate time during the planning and design stages. It will pay dividends down the line.
Glass: Parties can start to protect their position in relation to technology disputes from the very earliest stages of a contractual relationship. Although parties are naturally looking forward to a positive and successful working relationship, care should nonetheless be taken with the negotiation and drafting of clauses, which can make a great deal of difference to the smooth resolution of a dispute – such as a dispute resolution process – or the leverage available in such disputes, such as limitations of liability and limitation of actions provisions. During the life of an IT project, the key advice I would offer is to keep an eye to the provisions of the contract and to use them advantageously. A lot can be gained by using contractual mechanics, such as relief notices and change procedures, to protect a company’s position, and by using interim dispute resolution provisions to bring parties to the negotiating table at an early juncture.
Pack: From both the supplier and customer side, consider confidentiality in dispute processes. In England, court jurisdiction overwhelmingly means that a trial will be in public, if it gets that far. If a company wants to have confidentiality protections for sensitive information, such as trade secrets in the event of a dispute, it is important to have considered this at the time the contract was negotiated. However, the publicity and reputational impact of a public dispute can be valuable leverage that an opponent may not easily relinquish once a dispute has arisen. Early planning is key. Always remember that inaction when faced with a dispute is rarely going to end well. Even clauses in contracts which say that inaction is no waiver of a right or claim are not an absolute protection – doing nothing or failing to take action at the right time may lead to a loss of rights entirely. Even if it does not, there is the prospect of a potentially lengthy dispute just to determine whether or not rights have been waived, before getting into the true dispute. So, always act – or more accurately, decide whether to act or not – promptly.
Rachel Glass is a senior associate specialising in commercial litigation and dispute resolution with a focus on telecoms and technology disputes. She has co-authored the guide ‘IT Contracts and Dispute Management’, drawing from her 18 years of experience. She frequently advises on complex technology project disputes and has appeared in the Commercial Court, Technology and Construction Court and international arbitration. She can be contacted on +44 (0)20 7415 6733 or by email: rachel.glass@twobirds.com.
Jeremy Sharman is a partner in Bird & Bird’s London dispute resolution team with extensive experience advising on commercial disputes, focusing on the technology, communications and financial services sectors. He specialises in complex disputes and risk management issues related to the development and deployment of software solutions, the delivery and implementation of business-critical systems, and challenges relating to the award of public contracts for technology services. He can be contacted on +44 (0)20 7905 6214 or by email: jeremy.sharman@twobirds.com.
Matthew Pack’s practice focuses on technology implementation and outsourcing disputes, including contentious software deployments and failed legacy systems integrations. Alongside his technology focus, he is a member of Bird & Bird’s retail & consumer sector group, with experience of B2B and B2C disputes for retail businesses. He has extensive experience representing clients in high-stakes litigation, including in the English senior courts, in cross-border cases and international arbitration. He can be contacted on +44 (0)20 7415 6621 or by email: matthew.pack@twobirds.com.
Anne-Sophie Lampe is a partner in Bird & Bird’s Paris office, specialising in IT and internet litigation, as well as in IP issues when they relate to technology. With 15 years of experience in the technology sector, she advises tech companies and end users. She is passionate about digital transformation and AI, in particular generative AI-related challenges. She also teaches and writes extensively on digital law and IT-related topics. She can be contacted on +33 (1) 4268 6333 or by email: anne-sophie.lampe@twobirds.com.
© Financier Worldwide
International dispute resolution
Choosing a forum for international commercial disputes: litigation, arbitration or mediation?
Factors to consider when mapping objectives and tactics in dispute resolution
Confidentiality of the arbitration process: myth or reality?
The UK Arbitration Act 2025: key changes and summary disposal
Rising risks: navigating the growing landscape of AI litigation in the UK