Two years on from its launch, Relativity's aiR for Review has shifted from a novelty to a tool that review teams are using on real matters. It is worth taking stock of where it works, where the cost-benefit makes sense, and where it still needs careful handling.
What aiR for Review actually does
aiR for Review uses a large language model, integrated into Relativity, to make relevance and issue-coding predictions against a defined set of review criteria. Unlike classic Active Learning, it does not require thousands of human-coded documents to begin producing useful output: a well-written set of instructions and a small validation sample is often enough to get a meaningful first pass. Each prediction comes with a rationale the reviewer can read.
Where it earns its place
Three use cases have become reasonably well established:
- Early case assessment. Running aiR against the corpus very early, on a draft of the relevance criteria, surfaces likely hot documents and gives the legal team something concrete to refine scope and strategy against.
- Issue tagging at scale. Multi-issue matters, where each document might touch several themes, are where the rationale-per-issue output is most valuable. Reviewers spend their time confirming or rejecting structured suggestions rather than starting from a blank tagging panel.
- QC and prioritisation. Even where humans do first-pass review, aiR is being used to flag documents whose human coding looks inconsistent with the surrounding pattern.
Where it still needs careful handling
It is not magic. A few practical points come up repeatedly:
- Instructions are the work. The quality of aiR's output is largely a function of how clearly relevance and issues are written. Vague criteria produce vague predictions, just as they do for human reviewers.
- Validation is non-negotiable. Defensibility still depends on measuring recall and precision against a human-coded sample, and being able to explain the methodology if challenged.
- Privilege remains sensitive. Most teams continue to apply traditional searches and human review for privilege, with aiR as an additional safety net rather than a replacement.
Wider context
aiR is not the only player. Reveal, DISCO and others have shipped comparable generative-AI features, and the EDRM has published practical guidance on validation in its EDRM model updates. The UK courts have not, to date, treated AI-assisted review as inherently more suspect than other technology-assisted review approaches, provided the methodology can be explained.
Bottom line
Generative-AI review tools earn their place where they let smaller review teams handle larger, more complex matters without sacrificing defensibility. They do not remove the need for a properly scoped collection, clear relevance criteria or human judgement on privilege and key documents - and matters where any of those are weak are not the matters to try them on first.
§ From the guide, latest version
CFL Benefits Of Early Case Assessment In EDiscoveryCOMPUTER FORENSICS LAB · E-DISCOVERY TEAM CFLELECTRONICEVIDENCESERIES · EARLY CASE ASSESSMENT Benefits of Early Case Assessment in e Discovery A Practical Guide to Understanding Evidence, Controlling Disclosure Costs and Developing Case Strategy Earlier Understand the evidence before the evidence controls the case. ALIGNEDPRACTICE · CFLAB. UK · E-DISCOVERY. UK
§ CONTENTS In this guide 01 Executive summary and the ECA Principle 02 ECA and EDA: two related disciplines, one central proposition 03 The traditional problem: collect everything and review later 04 ECA and the disclosure framework in England and Wales 05 The strategic benefits of ECA (twelve benefits) 06 What modern e Discovery technology adds to ECA 07 The modern ECA workflow: twelve stages in a feedback loop 08 The evidence universe 09 ECA can reveal what is missing: source architecture and alternative evidence paths 10 Case study: Pyrrho Investments v MWB Property (2016) 11 Case study: Triumph Controls v Primus International (2018) 12 Worked case study: Halden Systems v a departing sales director (hypothetical) 13 Cost illustration and the e Discovery cost funnel 14 Questions ECA should answer (lawyer's checklist of 54 questions) 15 ECA deliverables and sample templates 16 ECA dashboard concept (illustrative data only) 17 ECA and privilege 18 ECA and UK data protection 19 Fifteen ECA mistakes that can become expensive later 20 When digital forensics should enter the ECA process 21 Practical implementation plan: first 24 hours to the DRD 22 Question banks: client, opponent and provider 23 Model instruction wording 24 Red flags 25 Glossary 26 Conclusion: early knowledge changes the economics of disclosure; specialist support 27 References and bibliography 28 Disclaimer and jurisdiction a l notes
§ 01 · EXECUTIVE SUMMARY Executive summary THEECAPRINCIPLE "Do not begin by asking how quickly every potentially relevant byte can be collected. Begin by asking what the legal team needs to know, which evidence could answer those questions, where that evidence is likely to exist, and what proportionate method should be used to preserve, assess and collect it." Early Case Assessment (ECA) is the structured, early analysis of a dispute: its facts, legal issues, risks, likely liability, key witnesses, evidence requirements, probable cost and possible strategy. Early Data Assessment (EDA) is the early analysis of the electronically stored information itself: custodians, sources, volumes, dates, file types, communications, duplication, concepts, keywords and collection requirements. The EDRM Identification Standards treat the two as the twin components of identification. This guide argues that the principal value of ECA is not a smaller review population. It is knowledge, obtained early enough that strategy, scope, budget and preservation decisions can still be changed. QUESTION SHORT ANSWER FIVEPOINTSTOTAKEAWAY
