Bloomberg Law News has published a report, titled "Robot Arbitrators Spark Conflicts Over AI in Dispute Resolution," examining the mounting conflicts surrounding the use of artificial intelligence tools in the arbitration process. The report, which appeared on the Bloomberg Law News service, addresses one of the most contested questions now facing the international dispute resolution community: whether, and to what extent, machines may perform functions traditionally reserved for human arbitrators.

The headline itself signals the fault line. The term "robot arbitrators" — a shorthand that practitioners use with varying degrees of enthusiasm and alarm — captures the scenario in which AI systems assume or assist with decisional tasks in arbitration, rather than serving merely as research or drafting aids. The word "conflicts" in the report's framing points to the practical disputes now arising among parties, counsel, and institutions over when such use is permissible, whether it must be disclosed, and what happens when it goes undisclosed or wrong.

Why the Issue Matters Now

Generative AI tools have moved rapidly from novelty to standard equipment in legal practice. Arbitration, with its confidentiality, party autonomy, and limited appellate oversight, sits in a distinctive position. Parties can, in principle, agree to almost any procedure they choose — including, some have argued, the use of algorithmic assistance in reaching an award. Yet the same features that make arbitration flexible also make it harder to police: no public docket, no routine scrutiny of deliberations, and enforcement challenges that surface only years later, often in a court far from the seat of the arbitration.

The Bloomberg Law report situates itself squarely in this tension. The conflicts it describes arise at the intersection of three long-standing pillars of arbitral legitimacy: the arbitrator's personal duty to decide, the requirement of independence and impartiality, and the parties' right to a fair process. An AI system does not owe duties, cannot be impartial in the human sense, and cannot be held accountable in the way an arbitrator can — through challenge, disqualification, or reputational sanction. That mismatch, the report's framing suggests, is where the disputes are concentrating.

The Practical Stakes for Practitioners

For counsel and arbitrators, the implications fall into several concrete areas. The first is disclosure. Institutions and drafters of arbitration rules are actively debating whether an arbitrator's use of AI tools must be disclosed to the parties, in the same way that relationships or prior engagements must be. The second is consent. If a party has not agreed to algorithmic involvement in the decisional process, an arbitrator's undisclosed reliance on such tools may expose the resulting award to challenge — whether at the seat, under the limited grounds for setting aside awards, or at the enforcement stage, where due process objections can be raised before enforcing courts.

The third area is the integrity of the record. Arbitral awards reasoned by AI systems raise the question of whether the reasons are the arbitrator's own. Several arbitral institutions have begun issuing guidance on generative AI, and the direction of travel, according to coverage such as the Bloomberg Law report, is toward explicit rules rather than silence. Practitioners advising clients on arbitration clauses should consider whether to address AI use expressly — both for arbitrators and for counsel — because an express clause today is cheaper than a challenge tomorrow.

A Divided Profession

The "conflicts" in the report's title also describe a profession that has not reached consensus. Some practitioners see efficiency gains: faster document review, quicker identification of controlling authorities, and lower costs that could make arbitration accessible to smaller disputes. Others see an existential threat to the fiduciary character of adjudication. The arbitrator's oath-like function — to hear the parties, weigh the evidence, and decide by reasoned judgment — does not translate neatly into a pipeline. The dispute resolution field is now working out, in real time, where to draw the line between assistance and substitution.

That line-drawing exercise has a familiar shape for anyone who has followed technology regulation: the tools arrived first, the norms are arriving second, and the case law will arrive third. The Bloomberg Law report contributes to the second phase by documenting where the friction has already surfaced.

What Comes Next

Three developments bear watching. First, institutional rules: arbitral bodies worldwide are reviewing or revising their guidance on AI, and a divergence between institutions would give parties another factor to weigh when selecting rules. Second, disclosure practice: early challenges and remissions involving undisclosed AI use will test whether existing standards — such as duties of independence, impartiality, and diligence — stretch far enough to cover algorithmic tools, or whether new express rules are required. Third, enforcement jurisprudence: national courts asked to enforce or set aside awards with an AI dimension will have to reconcile technology-neutral statutes with technology-specific concerns.

For now, the operative lesson for practitioners is caution and candor. Counsel drafting clauses, arbitrators considering tools, and parties evaluating institutions should all proceed on the assumption that AI use in arbitration will be scrutinized, and that silence in the arbitration agreement is not a strategy. The Bloomberg Law report's central insight — that AI in dispute resolution is generating conflicts, not consensus — is likely to remain accurate for some time.

This article is based on the Bloomberg Law News report "Robot Arbitrators Spark Conflicts Over AI in Dispute Resolution." Readers requiring the report's full contents, including named cases, quoted commentary, and specific rule references, should consult the original publication directly.