Kluwer Arbitration Blog

On the second day of Paris Arbitration Week (“PAW”) 2025, a sneak preview of the forthcoming  2025 International Arbitration Survey “The Path Forward: Realities and Opportunities in Arbitration” (“the 2025 Survey”) was presented at an event hosted at White & Case LLP (“W&C”) in Paris. The 2025 Survey is the fourteenth empirical project of Queen Mary University of London’s (“QMUL”) School of International Arbitration (“SIA”).

The 2025 Survey collates a wide range of views from every corner of the global international arbitration community. This year, the 2025 Survey involved 117 interviews and received over 2,400 responses.

Some preliminary findings of the 2025 Survey were discussed by a panel moderated by Clare Connellan (W&C). The speakers were Alexander Fessas (International Chamber of Commerce (“ICC”) International Court of Arbitration), Dr. Remy Gerbay (QMUL, SIA), Charles Nairac (W&C), and Aditya Singh (W&C).

 

Methodology of the 2025 Survey

Remy Gerbay kicked off the discussion by providing an overview of the empirical work of SIA and the methodology of the 2025 Survey. SIA/QMUL have two decades of experience in empirical research. The 2025 Survey is the sixth one in partnership with W&C.

The 2025 Survey was conducted by QMUL/SIA, under the academic lead of Norah Gallagher and Dr Maria Fanou, with the assistance of Dr Thomas Lehmann (W&C Postdoctoral Research Associate at QMUL).  As has been the case with all previous iterations of this type of empirical work, the 2025 Survey took an inclusive approach aiming to capture the views from as wide a range of arbitration stakeholders as possible (see here for previous coverage on the Blog of the methodology for the 2021 Survey).

The 2025 Survey was conducted by QMUL/SIA in two phases. The quantitative phase included the preparation of the survey questionnaire. The questions concerned, aside from the general recurring questions on preferences of practitioners (such as their preferred seats of arbitration and arbitration rules), important themes: sanctions, enforcement of arbitral awards, enhancing the efficiency and effectiveness of arbitration, the role of public interest and transparency, as well as the use of Artificial Intelligence (“AI”).

 

Experiences and Preferences

Perhaps unsurprisingly, 87% of respondents said arbitration was their preferred method of alternative dispute resolution. However, compared to the 2021 Survey (see also here), there was a notable decrease (by approximately 11%), in support for combining arbitration with another dispute resolution method.

Gerbay suggested the drop may be attributable to the shift back to business as usual as the world transitions away from COVID. Meanwhile, Alexander Fessas suggested that this might be due to parties having poor experiences with multi-tier clauses in their arbitration agreements.

The panel also discussed the respondents’ preferences by region. Each region largely retained a preference for its own seats. Globally, the five most preferred seats were London (43%), Singapore (31%), Hong Kong (31%), Beijing (20%) and Paris (19%). Panellists pointed out that the question was about preferences, rather than actual usage.

Several panellists noted the prevalence of Asian seats in these statistics, reflecting the preferences of the Asia-Pacific (“APAC”) respondents, who made up 47% of the total respondents. The panel also examined the respondents’ preferences for different arbitral rulesets. Each region appeared to prefer locally developed rules. The ICC rules featured in the top five most-preferred rules in almost all regions. Aditya Singh pointed out that globally, the Hong Kong International Arbitration Centre (“HKIAC”) ranked as the second most preferred arbitral ruleset, but it only featured in the top five most preferred rules in the APAC region.

 

Should Awards Set Aside at the Seat be Enforced Elsewhere? Minority Support for the Transnational Approach

The 2025 Survey asked respondents to answer ‘yes’ or ‘no’ to whether awards set aside at the seat should be enforceable elsewhere.  39% of respondents agreed with the proposition, favouring the transnational approach, whereas the remainder leaned towards the territorial approach. Despite being in the minority, Gerbay was enthusiastic about the support for the transnational approach. He claimed that this was a reaction to a real need to support enforcement efforts. Support for this view was strongest in South America (52% in favour) and weakest in the APAC region (64% against). Interestingly, the 2025 Survey also showed that 65% of junior practitioners were against the proposition, while only 55% of senior practitioners disagreed.

 

Room for Greater Efficiencies in Arbitration

The 2025 Survey asked respondents about how arbitration could be made more efficient. 49% of respondents supported the use of early determination, and 50% supported expedited procedures. The panel noted that these options would work to mitigate the due process paranoia which is common amongst arbitral tribunals.

Respondents also noted several issues that lead to inefficiencies in arbitration. Popular complaints included overly adversarial conduct by counsel, lack of proactive case management, and ‘over-lawyering’ – such as voluminous submissions and excessive document requests. Singh suggested that this showed that responsibility for these issues is shared between tribunals and counsel, and that these findings might encourage tribunals to be stricter with deadlines. Charles Nairac pointed out that while lawyers certainly have a role to play in improving arbitral efficiency, there is also an important place for clients in that discussion, who have significant influence over the initial content of an arbitration agreement.

 

Transparency of Arbitration: Respondents Continue to Appreciate Confidentiality

The 2025 Survey showed that 90% of respondents supported the continued confidentiality of commercial arbitration hearings. On the other hand, only 59% of respondents supported the publication of investments awards. Gerbay expressed surprise that this number was not higher, given the quasi-case law status that investment awards enjoy.

The panel was not surprised at the desire for confidentiality when arbitrating commercial matters that are not issues of public interest. However, Nairac questioned whether the arbitration community had learned from previous backlashes against arbitration, such as the proposed inclusion of arbitration in the ultimately doomed  Transatlantic Trade and Investment Partnership (“TTIP Agreement”). The risk of a scandal stemming from a confidential award rendered against the public interest should not be overlooked by the arbitration community.

The panel noted that confidentiality can create a risk to arbitral legitimacy, particularly where public interest issues are being discussed. However, while transparency remains a live issue, it was noted that awards are becoming increasingly accessible through online databases.

 

AI: More and More Use for Non-Decisional Tasks

Finally, the panel discussed the use of AI in arbitration. 90% of respondents expected to use AI at some point in the future, with the most common reason for using AI being to save time. 77% of respondents agreed that AI should not be used for making decisions.

Singh pointed to significant concerns surrounding the reliability and potential biases of AI tools. There is also a looming anxiety over how the use of AI tools will transform legal roles, particularly junior roles which may need to pivot to be focused on AI tool management, or be replaced by AI completely. Finally, the continued integration of AI tools into the legal sector may lead to a divide between firms and parties that can afford to invest in AI, and those that cannot.

Fessas suggested that the 2025 Survey shows what the real issue is: effective regulation. The European Union (“EU”) has recently passed laws regulating the use of AI (see here for our previous coverage of the impact of the EU AI Act), which sets out that legal decision-making is a high-risk activity that should remain human-driven and not be replaced by AI tools (see here for previous coverage on the Blog). On the other hand, ancillary procedures may be carried out by AI tools. The full impact of this new legislation on arbitration will need to be better understood, particularly as other countries may adopt similar laws. The arbitration community may not have exactly the same understanding of what constitutes an ‘ancillary procedure’ as the EU legislators.

While Gerbay agreed that regulation was important, he suggested that focusing on the distinction between decisional and non-decisional tasks may not be the best approach. What is most important is informed consent of the parties. Decision-making is a collective effort: judges rely on their clerks and tribunals rely on their secretaries. What legitimises this is informed consent. That is why understanding AI is necessary, so that parties can provide their informed consent to its use in any aspect of the arbitration process. Gerbay noted that it was well known that tribunal secretaries can be very involved in the award drafting process, and suggested that this may not be so different from using AI. Gerbay noted that technology was already used to resolve simple, low value disputes, and that it may only be a matter of time before AI tools fill this role.

 

Conclusion

The very-well attended event was a highlight of the 2025 Paris Arbitration Week. The final full report of the 2025 Survey will be published in the coming months. It was the general sentiment that, as is always the case with academic work, there will be fruitful debates on the findings of the 2025 Survey when they are released.

 

This post is part of Kluwer Arbitration Blog’s coverage of Paris Arbitration Week 2025.

________________________

To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.