"There is now a broad consensus that Artificial Intelligence (AI) research is progressing steadily, and that its impact on society is likely to increase…. Because of the great potential of Artificial Intelligence, it is important to research how to reap its benefits while avoiding potential pitfalls."
- Stephen Hawking, Research Priorities for Robust and BeneficialArtificial Intelligence, an open letter, 2015
Alan Turing, the renowned computer scientist, philosopher and the father of Artificial Intelligence (AI), developed the ‘Turing Test’ in 1950. It was a test to determine a machine's ability to exhibit intelligent behaviour equivalent to that of a human.
Today, 70 years later, AI is blurring the lines in one of the most complex areas of human activity - judging. AI is manning smart internet courts, receiving and analyzing pleadings and evidence, and delivering verdicts.
Artificial Intelligence Courts
Hangzhou, a city in north China, powers the country's cutting edge technological revolution. It is here that the first justice delivery system run by AI was introduced in 2017. Beijing and Guangzhou quickly followed suit. The three AI Internet Courts in China are judging disputes relating to online transactions of sale of goods and services, copyright and trademark, ownership and infringement of domains, trade disputes, and e-commerce product liability claims.
The Hangzhou Internet Court operates 24 hours a day and 7 days a week. The average duration of these online trials in Hangzhou was 28 minutes, and the average processing period from filing to trial and conclusion by a verdict was 38 days. The Hangzhou Internet Court is technically supported by Alibaba and its subsidiaries. This has raised serious concerns of both independence and impartiality.
Reliance on Block Chain Evidence
Cases dealt with by AI courts rely heavily on blockchain evidence. For the uninitiated, blockchain is literally a chain of digital blocks. It is the system of storing digital information (the block) in a public database (the chain). Blockchain preserves information about transactions like the date, time and purchase amount etc. A classic illustration would be a purchase on Amazon. It contains a series of transactions which are recorded and kept on a digital platform. Each ‘block’ added to the ‘chain’ comes into the public domain, where it remains preserved.
The critical question is, is blockchain tamper-proof? Is alteration of its data impossible by human intervention? Is blockchain data immutable and time-stamped, and can it safely be used as an auditable trail? The judges in China think so.
China’s Supreme People’s Court has put matters to rest. It has ruled that evidence authenticated with blockchain technology is binding in legal disputes. It ruled,
"...internet courts shall recognize digital data that are submitted as evidence if relevant parties collected and stored these data via blockchain with digital signatures, reliable timestamps and hash value verification or via a digital deposition platform, and can prove the authenticity of such technology used."
Nature of the Proceedings
The litigation process is conducted solely online, including the service of legal documents, the presentation of evidence, and the actual trial itself. To comply with the standards of a legal trial, it opts for an ‘in person and direct speech principle’ through an the online video system.
The Hangzhou Internet Court has on the virtual platform a black-robed AI judge. The AI Judge interacts with the litigants by video chat. Litigants (real) appear while an AI Judge (virtual) prompts them to present their cases. The black-robed AI Judge asks in a virtual pre-trial meeting:
“Does the defendant have any objection to the nature of the judicial block chain evidence submitted by the plaintiff?”
“No objection,” the human defendant answers.
What have China's AI Courts achieved?
China's official Xinhua news agency reported that more than 3.1 million Chinese litigation activities which took place from March to October 2019 were settled through the AI-powered smart internet courts. According to a report released by the Supreme People's Court, more than 1 million citizens are already registered with the smart court system, along with 73,200 lawyers. China’s has an internet penetration of 64.5% and has approximately 904 million internet users.
In December 2019, China’s Supreme People’s Court issued a white paper titled ‘Chinese Courts and Internet Judiciary’. In its policy paper, the Court endorses the use of:
“…big data, cloud computing, blockchain and artificial intelligence, the application of modules such as voice recognition in hearings, digitalized evidence presentation, automatic document verification, and simultaneous generation of e-files, intelligence-assisted case handling, and case management are all included in the toolkits for the judiciary”.
Time a critical factor in implementation
Time is a critical factor in today’s modern digital age. AI judging is giving solutions to an impatient generation which no longer has the time for the endless churning of the wheels of justice. But questions remain. Can Artificial Intelligence judges match up to human judging? Or will AI be more consistent?
In 1950, Alan Turing posed the fundamental question, "can computers think?" Almost seven decades later, Stephen Hawking had no doubts. In fact, he had apocalyptic warnings of an evolving superior artificial intelligence.
"It would take off on its own, and re-design itself at an ever increasing rate...humans, who are limited by slow biological evolution, couldn't compete, and would be superseded."
Future OF Artificial Intelligence in judging
Ever since the COVID-19 pandemic, the United States, the United Kingdom, Singapore, Taiwan, and many technologically advanced jurisdictions are already in the game. Several jurisdictions have commenced the courts under the new normal with the help of the current technologies. Though China’s use of AI is a precursor, the others are in the process of following suit.
Serious existential questions on AI judging are confronting justice administrations across the globe. Sensing the inevitability of AI in justice administration, the European Commission has adopted the ‘Ethical Principles Relating to the Use of Artificial Intelligence (AI) in Judicial Systems’.
Today, it is difficult to predict the future trajectory of Artificial Intelligence’s outreach in the justice delivery system. But to deal with the tomorrow, we need to start asking the questions now.
The author is a Senior Advocate practicing at the Supreme Court of India.