On May 29, 2020, I was invited by the Centre for Financial Regulation and Economic Development (CFRED), a research center of the Law Faculty of the Chinese University of Hong Kong (CUHK) to present a webinar on “Machine learning and the challenges of digital transformation in the law.” Over 70 attendees joined the webinar, including legal practitioners, academics, and students from Hong Kong, Mainland China, Singapore, and U.K.
Highlights from the webinar
In developing a healthy future of law where A.I. plays a key role, we need to promote better cross-disciplinary community understanding of:
- Concepts, value, and objectives of automation, machine-learning and analytics
- Technical and non-technical challenges of A.I. implementation: Data quality & volume, sensitivity, etc.
- Legal sector specific challenges: Different stakeholders with different capacities to own innovation processes
- Legal vs. statistical acceptance of ML, DL models
- A.I. bias, transparency and interpretability
- Raising tech baseline, supporting SME firms’ tech adoption
- Develop building blocks of legal A.I. “public goods”
The adoption of machine-learning (a form of artificial intelligence) by lawyers is shaping the future of law –as a profession and institution. Today, pockets of practice and traditional workflows have undergone massive transformation, viz. legal research and document reviews in discovery, regulatory disclosures, M&A, and contract management. Many alternative legal services providers, “legaltech” vendors, and other “NewLaw” businesses were born in the 2010s. Yet, truly full-service “robo-lawyers” are nowhere in sight. This seminar will examine –from a legaltech industry perspective– the technological influences on emerging industry trends, including the applications of analytics and natural language processing. The discussion will focus on how software encodes the law and legal processes in software, the related technological and ethical challenges, and the adoption barriers in legal practice.
Professor David Donald hosted the session and postdoctoral research fellow, Dr. Mahdi H. Miraz, moderated, while machine learning expert Mr. Tal Perry discussed deep learning applications – and even fielded a question about the new GPT-2 framework (see e.g. http://www.thisworddoesnotexist.com).
Overall, a great discussion on machine learning and natural language processing and their related research fields, computational linguistics, legal informatics, and ontology.