Reprinted with permission from the January 29, 2025 edition of the ALM Daily Report. © 2024 ALM Global Properties, LLC. All rights reserved. Further duplication without permission is prohibited, contact 877-256-2472 or asset-and-logo-licensing@alm.com.

 

Douglas J. Witten[1]

In 2011, IBM’s Watson computer system – then an early, noteworthy triumph of artificial intelligence (AI) – was making headlines. Watson had defeated a couple of former Jeopardy!  champions, dazzling with its encyclopedic knowledge and recall. Watson tantalized us with visions of applying AI technology to solve problems in groundbreaking ways, in the healthcare context and beyond.

Only that didn’t happened as planned. By 2022, Watson Health had fizzled and IBM sold it off to a private equity firm.[2]

Despite failing to reach the success for which IBM had hoped, Watson gave alternative dispute resolution (ADR) professionals an early AI “aha!” moment. With such capacity and promise, was AI next coming for our jobs as attorneys, mediators and arbitrators?[3]

Flash-forward to 2025, when AI has jolted the ADR field with its disruptive potential. The buzz around AI is deafening, and with good reason: Any competent ADR practitioner must understand the risks and rewards these new technologies bring to our field.

With 2011 but a distant memory, AI is poised to impact ADR substantially. Nonetheless, even as AI continues its inevitable advance into mediation and arbitration realms, key caveats remain.

Laying the Groundwork for Innovation

By 2020, the concept of online dispute resolution (ODR) had been around for years. However, ADR’s adoption and embrace of technology-aided mediation and arbitration, particularly via online video platforms, was not widespread before the implementation of COVID-19 lockdowns and travel restrictions.

Today, negotiating and resolving conflict remotely seems natural. Increasingly tech-savvy, we effortlessly interface with website chatbots, casually problem-solve with Alexa and Siri, and now consult with AI on any challenge we can dream up.

Adapting to ODR options has underscored the benefits technology brings to our ADR practices. Meanwhile, we more fully appreciate the tradeoffs inherent in balancing convenience, efficiency, and sound processes.

AI’s Meteoric Development

As ChatGPT and other large language models (LLMs), such as Gemini, Claude, and even legal-specific platforms have emerged, AI’s potential to shape ADR processes is eminently clearer. AI models’ capacity to cull and synthesize enormous sets of data has improved exponentially, engendering powerful new applications.

AI can support ADR providers in expediting logistical and administrative tasks, like scheduling and client communication, billing, marketing, and the circulation of standard forms. The American Arbitration Association (AAA) offers an AI chatbot to facilitate parties’ arbitration clause-drafting[4] – instead of repurposing boilerplate language, attorneys can use accessible AI tools to tailor clauses – and JAMS offers AI-assisted reporting services.[5] Examples like this will continue to proliferate.

AI can also enhance ADR proceedings themselves. Consider the mediator engaged with intransigent parties struggling to resolve a lawsuit. In preparation for the mediation – or even later into the process, out of ideas and facing the possibility of ongoing conflict or deadlock – the mediator might consult an AI platform to dissect risks and brainstorm negotiation strategies and potential solutions. Using AI to inject new perspectives based on decision science and analysis of likely judicial outcomes, for example, could mean the difference between resolution and protracted litigation.

Given the voluminous documents that might be involved in arbitration, and the need to analyze potentially complex facts and law, AI can also serve neutral decision-makers. Reviewing, organizing, and summarizing large volumes of documents, and creating initial agreement and award drafts, are other tasks for which AI is well suited.[6]

Balancing Risk and Reward

Incorporating AI tools, advocates and neutrals are wise to proceed cautiously, ever committed to upholding duties of professional responsibility, ethics, confidentiality, privacy, and security. Mediators and arbitrators must endeavor to limit bias and seek fairness in our ADR processes. Delegation to AI does not absolve practitioners of honoring these principles.

Practitioners must also understand how AI systems safeguard information and limit its accessibility to unauthorized viewers. Everyone has heard tales of so-called “hallucinations,” too, whereby an AI model generates misleading or incorrect results. Plainly, professionals incorporating AI into their ADR practices should be wary of trusting these tools without exercising due diligence and vetting them thoroughly.

Codes of professional responsibility, conduct, and ethics are increasingly emerging to address the AI explosion. Institutions including the American Bar Association,[7] AAA,[8] JAMS[9] and the Silicon Valley Arbitration & Mediation Center,[10] among others, have issued thoughtful rules and guidance to assist us in the AI-powered era.

AI Limits?

Mediation is typically less document-intensive than arbitration. As mediators lack decision-making power, they aren’t tasked with sorting and digesting high volumes of documents to make rulings or determinations.

And mediation is often about more than applying law to facts and reaching a “right” answer. Mediators routinely invoke nuanced skills like empathy, listening, and rapport-building to make parties comfortable engaging in the negotiation process. The raw logic of an AI system, without accompanying abilities to read parties’ emotions, body language, and vocal tone, does not supplant the human attributes mediators ply to resolve conflict.[11]

Arbitration requires neutrals to collect and sift through facts, draw legal conclusions, and decide cases. AI currently seems well-adapted to make certain decisions and logical deductions based on input data. However, for perhaps some of the same reasons we aren’t all yet in driver-less vehicles – even though the technology is here[12] – parties embroiled in high-stakes arbitrations for now remain unlikely to hand full decision-making authority to non-human, AI solutions.

Conclusion and a Look Ahead

Long before robot neutrals become commonplace, AI will serve as a collaborative partner that aids mediators and arbitrators in more efficiently performing complementary tasks. Employed conscientiously, AI can perform certain functions more efficiently than humans and, ideally, free neutrals to focus on exercising our uniquely human resolution skills.

Perhaps you are not quite ready to visit an AI-enabled “Dr. Watson” avatar instead of your family physician. But AI will continue to revolutionize best practices in the pharmaceutical and medical fields. Similarly, mediators and arbitrators who expand their capacities to resolve disputes by collaborating with AI tools will be the next ADR innovators.

As AI technology evolves, ADR practitioners owe it to our clients to remain educated and informed. The essence of ADR mechanisms will keep humans in the loop for the immediate future.

Yet AI presents immediate opportunities for economizing mediation and arbitration processes. A confluence of societal and attitudinal shifts, along with powerful technological advances, combine to stir a sea change. Appreciating AI’s potential to improve ADR practices, while also understanding its potential risks, is paramount to our ultimate goal: more satisfactory dispute resolution experiences.

[1] Douglas J. Witten (innovativeadr.com) is a mediator, arbitrator, and author who has 28+ years of legal experience.  He has mediated 1500+ cases since becoming a registered neutral in 2003 and offers dispute resolution services across areas including healthcare, commercial matters, injuries and accidents. Mr. Witten is Secretary of the State Bar of Georgia ADR Section and past Chair of the Atlanta Bar Association Dispute Resolution Section. He is author of The Stoic Negotiator™ newsletter and Time-Tested: The Stoic Negotiator’s Five Keys for the Modern Deal.
[2] See, e.g., Lizzie O’Leary, How IBM’s Watson Went From the Future of Health Care to Sold Off for Parts, SLATE (Jan. 31, 2022), https://slate.com/technology/2022/01/ibm-watson-health-failure-artificial-intelligence.html.
[3] Douglas J. Witten, Cyber-Mediators in ADR Future, DAILY REPORT (May 24, 2011), https://www.law.com/dailyreportonline/almID/1202551176665/. See also The Dark Side of Watson, NPR (Feb. 20,2011), https://www.npr.org/2011/02/20/133916058/the-dark-side-of-watson.
[4]4ClauseBuilder AI, AMERICAN ARBITRATION ASSOCIATION, https://clausebuilder.ai/ (last visited Jan. 21, 2025).
[5]JAMS PREVAIL TRANSCRIPTION, https://www.jamsadr.com/prevail-transcription (last visited Jan. 21, 2025).
[6]To wit, AAA has recently announced an AI-powered writing and document analysis tool. Bob Ambrogi, American Arbitration Association Partners with Clearbrief to Offer AI-Powered Legal Writing Tools to Its Panelists and Parties, LAWNEXT (Jan. 14, 2025), https://www.lawnext.com/2025/01/exclusive-american-arbitration-associationpartners-with-clearbrief-to-offer-ai-powered-legal-writing-tools-to-its-panelists-and-parties.html.
[7]Generative Artificial Intelligence Tools, AM. BAR ASS’N STANDING COMM. ON ETHICS & PROF’L RESP., Formal Op. 512 (Jul. 29, 2024), available at https://www.americanbar.org/content/dam/aba/administrative/professional_responsibility/ethics-opinions/abaformal-opinion-512.pdf.
[8]Principles Supporting the Use of AI in Alternative Dispute Resolution, AMERICAN ARBITRATION ASSOCIATION(Nov.2023),https://go.adr.org/rs/294-SFS-516/images/principles%20Supporting%20the%20Use%20of%20AI%20in%20Alternative%20Dispute%20Resolution.pdf.
[9]JAMS Artificial Intelligence Disputes Clause, Rules and Protective Order (effective June 14, 2024), available at https://www.jamsadr.com/files/uploads/documents/jams-ai-rules.pdf.
[10]Guidelines on the Use of Artificial Intelligence in Arbitration, SILICON VALLEY ARBITRATION & MEDIATION CENTER (Apr. 30, 2024), https://svamc.org/wp-content/uploads/SVAMC-AI-Guidelines-First-Edition.pdf.
[11]How might AI advance here? Some suggest that AI becomes more effective by highlighting the humans in AI instead of humanizing AI. See Oguz A. Acar, et al., Research: Consumers Don’t Want AI to Seem Human, HARVARD BUSINESS REVIEW (Jan. 7, 2025), https://hbr.org/2025/01/research-consumers-dont-want-ai-to-seemhuman.
[12]Lingering “glitches” are well-publicized but will surely become less frequent. See Laurie Perez, Dean Fioresi, LA Man Nearly Misses Flight as Self-Driving Waymo Taxi Drives Around Parking Lot in Circles, CBS NEWS (updated Jan. 5, 2025), https://www.cbsnews.com/losangeles/news/la-man-nearly-misses-flight-as-self-driving-waymo-taxidrives-around-parking-lot-in-circles/.