What happened
McConnell Dowell sued three insurers over a natural gas pipeline that it had built in Queensland. The construction contract and an arbitration had generated about 4 million documents, which the company cut to about 1.4 million. Justice Vickery found that traditional manual discovery was not likely to be cost effective or proportionate. At one minute per document, a junior solicitor would need over 23,000 hours, or over 583 working weeks. In September 2016 the court referred discovery to a Special Referee, who helped the parties agree on protocols. All parties agreed to use predictive coding. On 2 December 2016 the court adopted the referee's opinions and said that it would order TAR. The judgment says that TAR is likely to reduce the volume dramatically. It gives no figure for that reduction, and the review had not run.
Sources
-
Court document, by authors with no tie to the system.
Supreme Court of Victoria, Commercial Court, Technology, Engineering and Construction List (Vickery J). McConnell Dowell Constructors (Aust) Pty Ltd v Santam Ltd & Ors (No 1) [2016] VSC 734, 2 December 2016. AustLII copy, read through the Internet Archive.
McConnell Dowell has reduced the number of documents said to be relevant to the present proceeding from approximately 4 million to approximately 1,400,000. The use of predictive coding technology (TAR or Technology Assisted Review) is likely to dramatically reduce this volume and contain it further within reasonable and manageable bounds.
Details
AI system
- Workflow
-
The company scanned its documents into a searchable form and cut about 4 million to about 1,400,000. A Special Referee helped the parties agree on protocols for discovery. All parties agreed that predictive coding suited all of the documents.
Follow-up
- Conditions for replication
-
Questions of legal professional privilege wait until the predictive coding is complete. The parties confer on a protocol for the review. The case is large enough that manual review is not cost effective.
Entry details
- Sector
-
Legal services / e-discovery.
- Purpose
-
A vendor that the judgment does not name built the predictive coding software to classify documents as relevant or not, after learning from a sample that lawyers review.
Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Victorian court orders TAR for 1.4 million pipeline documents (AIBD-0832). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0832.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)