Meanwhile, Chinese courts are developing an AI system of “non-human judges” designed to offer comprehensive support, improve legal services, and strengthen justice across “smart courts” by next year.

Closer to home, former chief justice of India DY Chandrachud, a few days before his retirement on November 11, tested the intelligence of an AI “lawyer” by asking whether the death penalty was constitutional at the National Judicial Museum of the Supreme Court. The lawyer confirmed this, citing the “rarest of the rare” standard for heinous crimes, which visibly affected Chandrachud. In June, he advocated for a “moderate” adoption of AI in India’s judicial system.

Many countries have begun using artificial intelligence, and now generative artificial intelligence (GenAI) models, to reshape their legal systems, helping lawmakers, courts, and legal practitioners. From streamlining processes to predicting case outcomes, AI and law-specific language models promise to deliver efficiencies in many judicial systems and reduce the chronic delays and backlogs of millions of cases that plague courts around the world.

Goldman Sachs estimates that 44% of current legal job tasks could be automated by AI. Themis Solutions Inc. According to the 2024 Legal Trends Report by (Clio), 79% of legal professionals have adopted AI in some form, and a quarter use it widely or universally across law firms.

View Full Image

smart courts

Thousands of courts in China have mandatorily implemented AI-powered systems to help process cases and speed up routine decisions, significantly reducing processing times. People in China can use their smartphones to file complaints, follow the progress of a case, and contact judges. The country has also installed AI-based automated machines at so-called “one-stop” stations to provide 24-hour legal advice, register cases, create legal documents and even calculate legal costs. Judges and prosecutors use the Xiao Baogong Intelligent Sentence Prediction System in criminal law.

The Brazilian government is collaborating with OpenAI to accelerate the scanning and analysis of thousands of cases using artificial intelligence, aiming to prevent costly court losses that strain the federal budget. Brazil’s Ministry of Planning and Budget predicts that government spending on court-ordered payments will reach at least 100 billion reais (about 1% of the country’s GDP) in 2025. To reduce this burden, the Brazilian government is turning to artificial intelligence, especially to deal with small requests that collectively affect the budget but are difficult to manage individually.

The attorney general’s office (AGU) will apply artificial intelligence to prioritize cases, produce statistical analyzes for strategic planning, and summarize documents for court presentations. The AI ​​is intended to support AGU staff by increasing efficiency without replacing human workers who will oversee all output produced by the AI.

Tools such as LexisNexis and ROSS Intelligence (ROSS) can mine large libraries of case law, statutes, and precedents; These tasks are often tasks that will take days or even weeks for teams of lawyers. Judges and lawyers alike benefit from expedited progress, which allows them to focus on more nuanced aspects of cases.

As an example, Harvey is a GenAI platform specifically aimed at lawyers, built on OpenAI’s GPT-4. Its clients include PwC, and “more than 15,000 law firms” are on its waiting list. Looking more closely, companies like Lexlegis.AI, a Mumbai-based legal research company, and Sarvam, a Bengaluru-based developer of native language models, have developed law-specific large language models (LLMs) for the legal community in India.

Also Read: We need to reduce government lawsuits to unclog the judicial system

E-court project

While countries like India have yet to fully embrace AI in court decisions, the e-court project and other digitalization efforts are paving the way for potential AI integration in the country’s legal administration. For example, the vision document for phase 3 of the eCourts project states that “the framework will be forward-looking to include the use of artificial intelligence.”

“Courts and court systems have adapted to AI in some ways, but there is still a lot that can be done. For example, using AI to reduce the backlog. AI assistants or lawyers will essentially play the role of support teams. Senior advocate at the Supreme Court and Cyber ​​Saathi’ “But they can be used for pre-trial SWOT (strength, weakness, opportunity, threat) analysis,” said NS Nappinai, founder of .

“AI has not been implemented or tested beyond specific interventions in the Indian court system,” said Apar Gupta, advocate and co-founder of the Internet Freedom Foundation.

According to him, the India e-Courts committee project primarily focuses on digital transformation, addressing key issues such as computerizing court systems and facilitating remote litigation post-pandemic. As the judiciary primarily seeks to address structural challenges in infrastructure, staffing, and case processing efficiency, AI has been minimally applied and limited to tasks such as translating judgments into regional languages.

This is because, although courts around the world have recognized that AI can improve the efficiency and fairness of the legal system, the idea of ​​AI algorithms making “biased,” “opaque,” ​​and “hallucinatory” decisions can be deeply troubling.

According to Nappinai, various measures are being taken, but much more is needed. “First of all, although AI can be adapted, there will still be human intervention to oversee the results. The focus is now also shifting to cybersecurity requirements. Given the limitations of AI systems such as bias, hallucinations and incompetence, careful use of AI is adapting. Customized for India “We offer systems,” he added.

While simple automations like document watermarking and proofreading are used, “broader AI-based decisions require more careful and regular execution,” according to Gupta. “Generative AI (such as large language models, or LLMs) is viewed with caution as inherent inaccuracies can compromise fairness. While some initial interest in tools such as ChatGPT has emerged, judges remain largely cautious,” he added.

For example, this May, the Manipur high court took help of Google and ChatGPT to conduct research on service laws while deciding the writ petition of village defense force (VDF) member Md Zakir Hussain, who had moved the court. He will object to his “dismissal” by police officials for allegedly neglecting his duty.

And in March 2023, judge Anoop Chitkara of the Punjab and Haryana High Court used ChatGPT to obtain information in a bail hearing that involved ‘cruelty’ while committing a murder.

But five months later, Justice Pratibha M. Singh of the Delhi high court, while resolving a trademark dispute involving designer Christian Louboutin, ruled that GPT cannot be used by lawyers to provide reasons on “matters of law or fact in the court of law.”

Also Read: Generative artificial intelligence and its interaction with law

The United States has also used models such as COMPAS (corrective offender management profiling for alternative Sanctions) to predict the risk of recidivism (the tendency of criminals to reoffend) and influence bail, sentencing and parole decisions. However, this technology has faced serious criticism for perpetuating prejudices, especially against minority communities. In the Netherlands, welfare fraud detection AI SyRI encountered a setback; this system was discontinued due to racial profiling and privacy concerns.

To address such concerns, UNESCO has partnered with international experts to develop draft guidelines on the use of AI in courts. These guidelines, supported by UNESCO’s Recommendation on the Ethics of Artificial Intelligence, aim to ensure that AI technologies are integrated into judicial systems in a way that supports justice, human rights and the rule of law.

Increased impacts and risks

U.S. chief justice John G. Roberts Jr. warned of the growing influence of artificial intelligence in the legal profession in his 2023 year-end report, calling it the “cutting edge technological frontier.” He noted that AI could soon make traditional legal research “unimaginable” without its help, but also warned of its risks, including invasion of privacy and the risk of “dehumanizing the law.”

He highlighted the potential pitfalls of using AI in this area, citing a recent incident where lawyers who relied on ChatGPT were fined for citing non-existent legal cases. “Legal determinations often involve gray areas that still require human judgment,” Roberts said, among other things.

The document ‘Guidance on the Use of Artificial Intelligence in Canadian Courts’, published in September, acknowledges that while some judges in Canada have already adopted AI tools to improve their efficiency and accuracy, others may be unknowingly using generative AI. “Even if AI output is proven to be accurate and valuable, its use, especially when it comes to certain generative models, could unintentionally expose judges to legal complexities such as copyright infringement,” the report warns.

“What we need now is for court systems to adapt to technology to ease the burden of technology and streamline process-oriented aspects. It is critical that India acknowledges the positive aspects of technology use and overcomes resistance or fear in adapting to technology but in a cautious manner. They ( “Law-specific Masters) can be effective support tools, but they cannot replace human discretion,” Nappinai said.

Gupta, on the other hand, recommends integrating AI into legal practices with guidance from state bar councils and the Bar Council of India to help lawyers use generative AI “responsibly and effectively.” To leverage the efficiency of AI, he believes lawyers can use tools for specific tasks, such as case summarization, but must apply critical thinking to the insights generated by AI.

“For AI to positively transform legal practice, balanced regulation, ongoing education and careful implementation are essential, rather than rushing into AI as a blanket solution,” Gupta said.

Also Read: We need judicial system reforms to ensure that cases are concluded quickly