Paper presented at the AfAA 2nd Annual International Arbitration Conference, 15th – 16th April 2021

Executive Summary

Through the course of arbitral proceedings, practitioners can use an array of different technological tools to assist and facilitate their work. These range from tools to more efficiently sort documents through predictive coding, to artificial intelligence services which can predict costs and even outcomes of cases. New types of contracts such as smart contracts also utilize new types of arbitration proceedings such as Blockchain arbitration. 

Introduction

Modern technology has played and will continue to play a key role in the reform and innovation of the international arbitration practice. We have long grown accustomed to the ubiquitous use of e-mails, online legal research platforms and digital file storage. However, while modern technology has much more to offer, the arbitration community has historically been hesitant to explore and incorporate new technological developments into its daily practice.

This article follows up on the panel discussion during the African Arbitration Associations’ second Annual International Arbitration Conference (14 – 16 April 2021). It aims to provide a brief introduction on how technology is used in arbitration and give an outlook on technologies which may reform the international arbitration practice in the future.

Predictive Coding to Process Documents

At the outset of arbitral proceedings or prior to their initiation, parties may wish to use predictive coding to review large sets of documents and to determine the evidentiary base for their case.

The process of predictive coding starts by setting parameters and identifying documents which will form the sample set of documents to be reviewed. Each sample is coded as “relevant” or “not relevant” by a lawyer with good knowledge of the case, which in turn trains a predictive formula (algorithm). The algorithm is then applied to review the entire set of documents and can be further improved by human review of additional sample documents. The use of predictive coding aims to increase efficiency and achieve a higher level of consistency of relevant documents.[1]

Predictive coding is also used to identify documents during the document production phase to automate the process of identifying documents falling under the document production request. However, it may be disputed whether the use of predictive coding in this context is accurate enough to be appropriate. Therefore, the use of predictive coding should be disclosed to enable both the tribunal and the other party to ask questions to satisfy themselves that predictive coding was used in an appropriate manner. This can help avoid procedural irregularities which may give grounds for a challenge.[2]

The extent to which predictive coding is currently used in arbitration is unclear. As of 2018, only 5% of the respondents to the Queen Mary Survey on the Evolution of International Arbitration indicated that they frequently use artificial intelligence, including data analytics and technology-assisted document review.[3]

Artificial Intelligence to Predict the Outcome of a Case

Once parties have established the evidentiary base of their case (e.g. with the help of predictive coding), they may again turn to Artificial Intelligence (AI) tools to gauge their chances of success prior to filing the case.

At its core, AI is a branch of computer science focused on replicating or simulating human intelligence in machines in order for them to think like humans and mimic their actions.[4] Predictive analytics as an AI tool are increasingly used to predict the outcome of disputes. These tools are particularly attractive to litigation funders, which aim to achieve a competitive advantage in their risk assessment. One example for an AI tool used by litigation funders is Legalist, which claims to use data from millions of court records to support the case assessment.[5]

In the arbitration context, Arbilex, founded in 2018, uses AI and predictive analytics to assess the likelihood of success of arbitration cases, as well as the likely costs.[6] However, the data set of available court cases is generally significantly larger than for awards, which impacts the accuracy of these AI tools. It remains to be seen whether arbitral institutions will allow AI platforms access to their awards.[7]

While the prediction of the outcome of a case generated by AI can be appealing to funders and practitioners alike, one must carefully assess the risk of bias embedded in AI. Since AI software is trained on large-scale data, if such data includes racial biases, these are incorporated into the training data. For example, the US program COMPAS assesses the likelihood of recidivism in defendants who are up for parole. While the algorithm makes no reference to ethnicity or race, a study has shown that the COMPAS assessment is biased against African Americans.[8] Thus, if predictive analytics and AI are to be used more widespread in arbitration it is paramount that the arbitral community institutes safeguarding mechanisms and checks to minimize the impact of bias of AI technologies.[9]

Online Case Management Platforms

Once a party decides to proceed to arbitration, be it because AI tools predicted a high chance of success or at its own accord, the party will face different levels of digitalisation depending on the arbitral rules applicable to its case.

First, a party will need to determine whether an arbitration can be commenced by filing the claim electronically rather than in hardcopy. Many institutions have now switched written submissions and communications to be predominantly electronic. Arbitral institutions also differ in whether they (for internal purposes) use fully electronic case management systems such as the Vienna International Arbitral Centre (VIAC) and the Stockholm Chamber of Commerce (SCC). However, a more recent development and a step towards fully digitalizing the administrative aspects of arbitration is the adoption of online case management platforms to which also the arbitrators and parties have access. The SCC and VIAC now also provide such online case management platforms.[10]   

Chinese arbitral institutions such as Shenzhen Court of International Arbitration (SCIA) and Guangzhou Arbitration Commission (GZAC) have also invested in digital platforms, which not only allow for online filing and processing payments but also offers a remote hearing platform. As a special feature of these platforms, the voice-to-text technology automatically generates and displays a transcript of the hearing synchronously during the hearing.[11]

Remote Hearings

The travel restrictions and social distancing rules imposed in many countries in reaction to the COVID-19 pandemic triggered a re-evaluation in the arbitration community of remote arbitral hearings. Several other procedural aspects were already regularly conducted remotely, such as organisational conferences and submissions via e-mail, but most hearings were still held as physical in-person hearings.

In general, national arbitration laws rarely address the question whether hearings can be held remotely, even if one party objects. Many modern and flexible institutional rules either expressly or implicitly allow hearings to be conducted remotely, such as Vienna International Arbitral Centre (VIAC) and London Court of International Arbitration (LCIA). It is then within the discretion of the arbitrators to evaluate whether a remote hearing can safeguard the parties’ right to be heard and the right to fair and equal treatment.[12]

It is increasingly accepted among arbitration practitioners that an oral and synchronous exchange of arguments or evidence in a remote hearing can, in principle, satisfy a party’s right to a hearing.[13] Conversely, not holding the hearing remotely when one or more parties cannot attend in-person for an extended period of time may violate the parties’ right to a hearing and the arbitrators’ obligation to conduct the proceedings efficiently and expeditiously.[14]

Experience gained, especially over the last year, has shown that well-organised remote hearings which properly adhere to recommended protocols and guidelines can be an efficient and comparatively low-cost method of conducting arbitrations.[15]

The increased use of semi- and fully remote hearings, where appropriate, could remedy the growing criticism of arbitration as being too expensive and too long. Furthermore, the use of remote hearings can address concerns of the environmental impact of international arbitral proceedings and foster an increase in diversity of arbitrators.[16]

Blockchain Arbitration

A comparatively new technology used for online dispute resolution is so called blockchain or “on chain” arbitration which has been developed primarily for disputes arising out of smart contracts.

Blockchain is a specific type of database, which differs from a typical database in the way data is stored. It is an incorruptible digital ledger of transactions which can be programmed to record not only financial transactions but almost anything which is of value to be recorded. The data is not stored in a single location but spread across a network of millions of computers simultaneously.[17] Blockchain is mostly used in connection with cryptocurrencies such as Bitcoin, but have numerous other applications.[18]

Smart contracts have arisen from blockchain technology and, unlike regular contracts, are not written in a specific language such as English but completely in code. Based on blockchain technology, smart contracts are self-executing, and transactions are trackable and irreversible.[19]

To serve disputes arising from smart contracts, blockchain arbitration was developed. Blockchain arbitration is structured as an automatic dispute resolution system, which is combined into a digital asset system. Arbitrators are provided with the ability to implement decisions directly on the blockchain or within the system, without the need for a paper award. The UK Jurisdiction Taskforce recently published draft rules for such dispute resolution mechanisms, expressly aimed at resolving disputes arising from new technologies such as cryptoassets, - currencies, smart contracts etc.[20]

One example for a blockchain arbitration system is the online dispute resolution platform Kleros, which was founded in 2017.[21] Kleros is a decentralized arbitration process relying on crowdsourced adjudicators, or “jurors”, who are randomly assigned to a case. After assessing the evidence, jurors commit their vote to one of the contractually specified options. The option with the most votes is the winning one. Each juror who is coherent with the final decision will then be paid a specified fee while the arbitrable smart contract defines which party has to pay the arbitration fee.[22] Jurors are financially incentivised to decide coherently. However, since deciding “coherently” is defined as deciding in line with the majority, it can be debated whether this incentivises correct decisions or rather a bet on how the majority would rule even if this is contrary to the individual juror’s evaluation, which may be based on special expertise.

Conclusion

The COVID-19 pandemic has forced the arbitral world to reconsider its scepticism towards technology and adapt quickly in order to continue serving its clients and secure the continuation of pending and new arbitral proceedings. Most notably, many practitioners and arbitral institutions were forced to predominantly work remotely – often for the first time.

It remains to be seen whether the momentum gained from the pandemic with regard to some technologies – such as online hearing platforms – will increase the acceptance of other technologies as well.

___________________________ 

*Attorney at Law, Binder Grösswang

[1] Claire Morel de Westgaver & Olivia Turner, Artificial Intelligence, A Drive for Efficiency In International Arbitration – How Predictive Coding Can Change Document Production, Kluwer Arbitration Blog (Feb 23, 2020), http://arbitrationblog.kluwerarbitration.com/2020/02/23/artificial-intelligence-a-driver-for-efficiency-in-international-arbitration-how-predictive-coding-can-change-document-production/.

[2] Claire Morel de Westgaver & Olivia Turner, supra.

[3] 2018 International Arbitration Survey: The Evolution of International Arbitration, Queen Mary University of London, http://www.arbitration.qmul.ac.uk/media/arbitration/docs/2018-International-Arbitration-Survey---The-Evolution-of-International-Arbitration-(2).PDF.

[4] Jake Frankenfield, Artificial Intelligence (AI), Investopedia (Mar 8, 2021), https://www.investopedia.com/terms/a/artificial-intelligence-ai.asp.

[5] https://www.legalist.com/philosophy.

[6] http://www.arbilex.co/welcome.

[7] Eric Chang, A Roundup of Tech and Dispute Resolution News, Kluwer Arbitration Blog (Apr 14, 2021), http://arbitrationblog.kluwerarbitration.com/2021/03/09/a-roundup-of-tech-and-dispute-resolution-news/..

[8] Id.

[9] Brand Rosen, Global Experts Keep it Real in Webinar Exploring Artificial Intelligence and its Role in Arbitrations and Legal Practice, Kluwer Arbitration Blog (Dec 12, 2020), http://arbitrationblog.kluwerarbitration.com/2020/12/12/global-experts-keep-it-real-in-webinar-exploring-artificial-intelligence-and-its-role-in-arbitrations-and-legal-practice/.

[10] Allison Goh, Digital Readiness Index for Arbitration Institutions: Challenges and Implications for Dispute Resolution under the Belt and Road Initiative, 38 J. Intl. Arb. 253, at 256-259 (2021).

[11] Id.

[12] Alice Fremuth-Wolf, Ingeborg Edel & Anna Förstel, How the COVID-19 pandemic may shape the future of international arbitral proceedings, IBA, Arbitration Committee Publications (Nov. 25, 2020), https://www.ibanet.org/Article/NewDetail.aspx?ArticleUid=A7F75D89-2CFD-4386-96B9-53341D0A55DA.

[13] Compare Maxi Scherer, Remote Hearings in International Arbitration: An Analytical Framework, 37 J. Intl. Arb. 407 (2020).

[14] Compare Kun Fan, The Impact of COVID-19 on the Administration of Justice, Kluwer Arbitration Blog (Jul 10, 2020),http://arbitrationblog.kluwerarbitration.com/2020/07/10/the-impact-of-covid-19-on-the-administration-of-justice.

[15] Id.

[16] Compare Lucy Greenwood & Kabir Duggal, The Green Pledge: No Talk, More Action, Kluwer Arbitration Blog (Mar 20, 2020), http://arbitrationblog.kluwerarbitration.com/2020/03/20/the-green-pledge-no-talk-more-action; Alice Fremuth-Wolf, Ingeborg Edel &Anna Förstel, supra.

[17] Derric Yeoh, Is Online Dispute Resolution The Future of Alternative Dispute Resolution?, Kluwer Arbitration Blog (Mar 29, 2018), http://arbitrationblog.kluwerarbitration.com/2018/03/29/online-dispute-resolution-future-alternative-dispute-resolution/.

[18] Luke Conway, Blockchain Explained, Investopedia (Nov 17, 2020), https://www.investopedia.com/terms/b/blockchain.asp.

[19] Jake Frankenfield, Smart Contracts, Investopedia (Mar 25, 2021), https://www.investopedia.com/terms/s/smart-contracts.asp.

[20] Eric Chang, supra.

[21] https.//kleros.io/.

[22] Clement Lesaege, Federico Ast & William George, Kleros, Short Paper v1.0.7, Sep. 2019, at 1 ff.