AI eDiscovery Hong Kong: Are AI Tools Essential for Legal Practice?

Author: Erica So, Associate Solicitor AI eDiscovery in Hong Kong is increasingly reshaping the way legal professionals approach document review and litigation workflows. While the legal industry has traditionally been conservative, advances in Artificial Intelligence (AI), Generative AI, and machine learning technologies are gradually transforming key processes, particularly in the discovery phase of litigation.

The legal industry, known to be profoundly traditional, is seeing slow but significant technological transformation, driven by advances in Artificial Intelligence (AI), Generative AI (GenAI), and other machine learning technologies. With a view to increase speed and accuracy in workflows especially within the discovery stage in litigation, tool-assisted review (TAR) has steadily gained in popularity in the past decade.

Nowadays, with the emergence of TAR 2.0, it is set to showcase the full range of AI’s potential and allow legal professionals to target only information that is essential and improve their overall efficiency in reviewing documents for court actions.

The Use of AI in e-Discovery

The use of AI in e-discovery since early 2010 has paved the development of TAR 2.0. The TAR workflow streamlines the documentation review process by predicting which documents are relevant based on patterns identified in a set of sample documents. The process involves harnessing machine learning technology to identify potentially relevant documents during discovery, a stage in litigation where both parties exchange information and evidence.

Practitioners begin by coding a set of sample documents, known as the “seed set”, which acts as a training dataset for the AI system. By categorising and annotating these documents, the system is trained to recognise relevant material.

The system then identifies similar documents and organises them according to relevance. This predictive coding assigns probability scores to determine whether documents are likely to be relevant. This significantly reduces the time and effort required to review large volumes of data and improves overall accuracy.

From TAR 1.0 to TAR 2.0

While TAR 1.0 has been effective in reducing manual review and lowering costs, it relies heavily on a static seed set. This means that:

• relevance decisions depend on initial data selection

• multiple rounds of manual review may be required

• accuracy may vary depending on dataset quality

This limitation has led to the development of TAR 2.0.

TAR 2.0 and Continuous Active Learning

TAR 2.0 uses Continuous Active Learning (CAL) models.

Like TAR 1.0, it begins with a seed set of documents. However, the system continues to learn and adapt dynamically as reviewers interact with it.

This allows:

• continuous refinement of document relevance

• increased accuracy over time

• improved efficiency across different cases

Importantly, each matter has its own dataset, meaning the system remains tailored to the specific facts and legal issues of each case.

Limitations of AI in Legal Automation

While AI tools offer greater speed and accuracy compared to manual review, limitations remain.

The CAL model primarily relies on text-based analysis, meaning:

• images, audio files, and videos are not easily analysed

• structured data such as spreadsheets may pose challenges

• contextual interpretation may still require human judgement

As a result, AI is best viewed as a supporting tool, rather than a replacement for legal professionals.

The development of fully automated legal systems remains ongoing, and the focus continues to be on improving accuracy and expanding the types of data that can be effectively reviewed.

How Ravenscroft & Schmierer Can Help?

Ravenscroft & Schmierer advises clients on litigation strategy and evolving practices such as AI eDiscovery Hong Kong, including the use of technology-assisted review tools. The firm supports clients in managing complex document review processes and aligning technology use with legal and procedural requirements. Clients seeking further information may contact us.

FAQ: AI eDiscovery Hong Kong

What is AI e-discovery?

AI e-discovery refers to the use of artificial intelligence to review and identify relevant documents during litigation.

What is tool-assisted review (TAR)?

TAR is a process that uses machine learning to analyse documents and identify relevant information.

What is the difference between TAR 1.0 and TAR 2.0?

TAR 2.0 uses continuous learning models, while TAR 1.0 relies on fixed training datasets.

Is AI more accurate than manual document review?

AI can improve efficiency and consistency, but human oversight remains important.

What are the limitations of AI in e-discovery?

AI may struggle with non-text data and complex contextual interpretation.

How can Ravenscroft & Schmierer assist with AI-driven litigation processes?

Ravenscroft & Schmierer supports clients in managing document review and litigation processes involving technological tools.

Why work with Ravenscroft & Schmierer on e-discovery matters?

Ravenscroft & Schmierer combines legal experience with awareness of evolving technologies to assist clients in handling complex litigation efficiently.

---

Disclaimer: Whilst every effort has been made to ensure the accuracy of this article, it is general in nature and does not constitute legal advice of any kind. You should seek your own personal legal advice before taking legal action. We accept no liability whatsoever for loss arising out of the use or misuse of this article.

For specific advice about your situation, please contact us.

Erica So

Associate Solicitor

+852 2388 3899