Care is required when using AI for case law analysis: Justice Prathiba Singh of Delhi High Court

Justices Prathiba M Singh and Rajiv Shakdher of the Delhi High Court on Saturday shared their perspectives on the intersection of technology and law at the Society of Indian Law firms (SILF) Artificial Intelligence (AI) conference..Both judges called for embracing technology while also exercising caution. Justice Singh said that laws need to change to cover the regulation of AI-related technologies.“If laws do not change, we might reach a point where laws become irrelevant. Because technology is on a marathon already. It is not going to stop,” she said..Drawing a distinction between an AI-assisted invention, an AI-generated invention and AI as a co-inventor, Justice Singh said, “Provisions as they exist doesn't permit AI to be an inventor. I believe there will be a time when law would need rethinking to accept AI as an inventor. But the question is, the idea of IP is to reward human creativity, the human spirit of creativity. It is a moral, ethical, legal question. If you say no to AI inventions, in the future all inventions will be refused patent. It is a hare and tortoise race. If you refuse patent to AI inventions, there is the threat of something being only a trade secret and will have us all rethink how the legislation will have to be.”While batting for AI inventions, Justice Singh cautioned against AI hallucinations and said care is required when using AI for case law analysis.Pointing out the limitations of AI, Justice Singh said that the accuracy and reliability of AI-generated data is still in the grey area.She said that AI cannot substitute either human intelligence or the human element in the adjudicatory process. At best, AI tools could be utilised for a preliminary understanding or for preliminary research and nothing more..Justice Shakdher stated that the current hullabaloo around technology was nothing but an attempt to replicate human beings and stressed the need for technology to be dominated by human beings."Technology needs to be dominated by human beings, because if you don’t do it, you would actually make something which is not in control of human beings,” he said.Sharing his perspective on AI and the opportunities and challenges it poses, he said, “We should look at the technology with caution and not give up on it.” Referring to the Mata case, the first case of AI hallucinations where two lawyers used ChatGPT to prepare a filing, but the bot delivered fake citations, Justice Shakdher said, “What you find in technology is a little less than what God would do. It has the strengths and weaknesses of a human being. Therefore, we see that the machine lies.” He urged lawyers to question how technology can serve and assist them without taking away their power of critical thinking. .Dr Rajiv Mani, Secretary of the Legislative Department, Ministry of Law & Justice, also addressed the conference. He emphasized that government policies and programmes are inspired by the spirit of 'AI for all' and, therefore, the use of AI should not be limited to the elite and technologically empowered. .During a technical session on Day 2 of the conference, Managing Partner of Saikrishna & Associates Saikrishna Rajagopal delved into how copyright in works generated via AI is treated in different jurisdictions.Partner at Brown, Neri, Smith & Khan, LLP Dr Ryan Abbott opined that the debate around AI and authorship was an “absolutely hot mess" particularly in the US, given the special interest in inventorship. .Another session on Artificial Intelligence and IP: Beyond Trade Related Aspects of Intellectual Property Right (TRIPS)' saw Senior Consultant (IPR), Department for Promotion of Industry and Internal Trade (DPIIT), Dr Raghavender GR saying,“Just like criminal laws and Patent laws are amended, we might have laws amended to make space for AI-assisted inventions.”He, however, junked the need for granting IP rights on AI-generated inventions or inventions co-created with AI.Panel moderator and Managing Partner at Anand and Anand Pravin Anand touched upon Indian courts granting legal rights to non-living entities like deities, forests, wildlife, etc and wondered if similar treatment could be meted out to a machine.To this, Dr Raghavender responded, “There should be no anthropomorphism, putting life into machines. Joint authorship is a bad idea.” .The final session of the conference was on Evolving International Legal and Regulatory AI Landscape; Challenges and Way Forward.Director and Advisor to Government of India for Global Partnership on AI Rohan Gupta said that regulations and governance differ from region to region.He explained that the developed world can look at privacy and other aspects around AI, but the global south will focus on developmental aspects of AI and their sectoral needs like healthcare, pharma, agriculture etc.“Right now, what we have seen is focus on safe and secure AI. Only China is talking about deployment of AI. The basic risk of AI remains the same overall. AI is borderless. But then each country builds upon it. Like what India did on Digital Personal Data Protection (DPDP). India is taking a lead, we have seen in G20,” he added.