Transcribing Court Proceedings with AI Technology: An Analysis

The Supreme Court of India has recently come up with the decision to make use of AI-powered natural language processing technologies in transcribing court proceedings. The idea is to capture what people inside the court were speaking and convert it from speech to text.

This intelligent automation will speed up the process of creating transcripts that are later made available to various stakeholders such as lawyers, parties involved in the concerned case, etc. Quicker access to transcripts will benefit lawyers, especially during multi-day hearings. This AI solution, named – Technology Enabled Resolution (“TERES”), developed by Bangalore-based Nomology Technology Private Limited, was already being used for transcribing arbitration matters and has proved its value because specialist transcribers often had to be hired from abroad, adding significantly to the overall cost borne by the parties.

With AI already permeating much of human society, it was only a matter of time before the judiciary also adopted it in the sphere of litigation. Nonetheless, the decision to move ahead with the experiment is laudable and especially so in light of the apex court’s recent decision allowing live-streaming of certain hearings (i.e., of only those hearings that relate to the interpretation of our constitution and where the bench comprises five or more judges).

These two steps will force all stakeholders in our legal ecosystem to change their mindsets, ways of working, and in-court behaviour. Over time, one hopes that such changes will collectively yield various benefits some of which are listed below:

  • Improved justice delivery system (based on better arguments and more efficient access to and assessment of evidence).
  • Reduced pendency of cases (based on faster disposal of matters as well as a decline in the tendency to approach courts for frivolous matters).
  • Easier and cost-effective access to legal recourse for larger sections of our society.
  • Minimised use of time-wasting tactics (e.g., needless adjournments).
  • Higher standards of courtcraft and a better understanding of the context in which certain comments are made by the bench or the bar.
  • Better recordkeeping.
  • More accountability.
  • Enhance learning for newer generations of lawyers.

For years, the government has sought to make India a preferred centre for international arbitration and mediation. Progress on the ground has however been slow. The wider use of modern technologies may prove to be a catalyst in this regard. This can also be a boost to improve the ease of doing business in India, especially at a time when, for various geopolitical and economic reasons, foreign direct investment is on the rise. At the same time, the burgeoning start-up scenario in India is attracting significant private equity and venture capital. A lot of intellectual property is being created in India and needs to be suitably protected. Especially because much of it has to do with emerging areas that are critical to the future of our country and indeed, the world.

To be sure, there will still be various practical challenges that need to be ironed out. As pointed out by the Hon’ble Chief Justice of India, Dhananjaya Y Chandrachud, multiple voices at the same time may well confuse the AI tool and hinder accurate transcription. Different accents and loudness of voices may also potentially complicate matters. Also, unless the entire judiciary (across all courts) adopts such technologies, the benefits will be limited. Such widespread adoption may still be derailed by objections from various quarters.

Therefore, it is too early to conclude with any level of certainty that the above-mentioned benefits (and possibly, others) will indeed be realized, and if yes, how long it will take. However, the Supreme Court’s decision to use technologies to usher in greater levels of efficiency and transparency is a clear signal of intent. As the old saying so presciently reminds us, even a journey of a thousand miles begins with the first step. That step has been taken.

Image Credits:

Photo by Tara Winstead: https://www.pexels.com/photo/clear-mannequin-on-dark-blue-background-8386365/

For years, the government has sought to make India a preferred centre for international arbitration and mediation. Progress on the ground has however been slow. The wider use of modern technologies may prove to be a catalyst in this regard. This can also be a boost to improve the ease of doing business in India, especially at a time when, for various geopolitical and economic reasons, foreign direct investment is on the rise. At the same time, the burgeoning start-up scenario in India is attracting significant private equity and venture capital. A lot of intellectual property is being created in India and needs to be suitably protected. Especially because much of it has to do with emerging areas that are critical to the future of our country and indeed, the world.

POST A COMMENT

The Curious Case of the Robolawyer (No, it's not a Perry Mason Novel!)

With the advent of technology, there is a drastic increase in the use of AI (Artificial Intelligence) which has significantly altered the way technology is perceived and will have a far-reaching impact in the future. Hence, it becomes necessary to try to minimize its shortcomings and make prudent use of the technology.

I do not know how many of you have heard of Joshua Browder, the 26-year-old founder of DoNotPay, a US-based venture that has developed a “robolawyer”- essentially an AI-powered bot that helps users in use cases such as appealing vehicle parking tickets, negotiating airline ticket refunds, and contesting service provider bills. Although the app was first released in 2015, to be honest, until recently, I too had not heard of him or the app!

My curiosity was piqued when I recently read the news that his company is willing to pay a million US dollars to any person or lawyer willing to repeat verbatim in front of the Supreme Court judge all that their robolawyer asks them to. It remains to be seen whether someone will take Josh up on that offer, whether the US Supreme Court will grant permission and what the outcome will be. However, it is being reported in the media that the DoNotPay app will help two defendants argue speeding tickets in US courts next month. The company has promised to pay the fines on behalf of the users if the robolawyer loses their appeals.

The app runs on the AI model known as “Generative Pre-trained Transformer” or GPT. This is the same technology that runs ChatGPT, which reportedly hit a million users in less than a week of its launch. AI technologies are constantly improving, and there is now greater emphasis on “ethics” and “explainability.” Essentially, the software must be able to explain how it arrived at a certain conclusion or output. This is important to minimize, if not altogether eliminate, the risk of biases and prejudices that creep into AI software simply because it is trained using hundreds of millions of content elements on the web (articles, images, reports, videos, etc.) that were all created by humans, and as such, carry the individual beliefs, prejudices, convictions, etc. of their original creators.

Over the coming decades, AI will shake up many fields including legal practice, healthcare, finance, etc. Not all fields will be impacted at the same pace or to the same extent but change they will. Already, AI is being used by healthcare professionals in improving the efficacy of diagnosis and confirmation of lines of treatment. Law firms too are beginning to use AI to simplify the tedium of the process of trawling through case laws and legal judgments to identify precedents and the reasoning of the benches involved. Soon, lawyers will simply be able to type in questions into ChatGPT, which will provide well-reasoned answers in a matter of minutes. Of course, the real skill will be to ask the right questions and figure out how sensible the answers are, and decide on further courses of action. Think of it as an advocate briefing a senior lawyer before the latter argues in court.

Half-baked knowledge is dangerous. For many years, patients (and/or caregivers) have used search engines to find information about symptoms, diagnostic tests, and lines of treatment and then argue with qualified medical professionals about their choices, at times forcing doctors to explain their hypotheses and reasoning. It is quite likely that in the foreseeable future, clients of lawyers and law firms too will be tempted to adopt a similar approach, which means lawyers too will end up spending time and effort on educating clients on matters of law and jurisprudence. Maybe it is worth coming up with new pricing models to dissuade frivolous “brainstorming” and “legal strategy” sessions!

Note to myself: Try out ChatGPT to explore the kind of responses it provides and start preparing for a future that will undoubtedly be more closely linked with AI tools.

References:

[1] Design Application Numbers 274917, 274918, 284680, 276736, 260403

[2] 24 U.S.P.Q.2d (BNA) 1614 (BPAI Apr. 2, 1992)

[3] Apple, Inc. v. Samsung Elecs. Co., 926 F. Supp. 2d 1100 (N.D. Cal. 2013) (partially affirming jury damages award).

[4] US6763497B1

[5] US10915243B2

Image Credits:

Photo by cottonbro studio: https://www.pexels.com/photo/person-using-macbook-3584994/

Over the coming decades, AI will shake up many fields including legal practice, healthcare, finance, etc. Not all fields will be impacted at the same pace or to the same extent but change they will. Already, AI is being used by healthcare professionals in improving the efficacy of diagnosis and confirmation of lines of treatment. Law firms too are beginning to use AI to simplify the tedium of the process of trawling through case laws and legal judgments to identify precedents and the reasoning of the benches involved.

POST A COMMENT