India’s judicial backlog is often reduced to a numbers problem: too many cases, too few judges, and years of waiting. But the numbers themselves show just how deep the problem is. More than 5.1 crore cases are currently pending across the courts covered by the National Judicial Data Grid (NJDG), with nearly 48 lakh cases pending for more than 10 years. Another 84 lakh have been pending for five to 10 years.
The capacity challenge is equally stark. As of today, India’s district and subordinate courts have a sanctioned strength of 25,894 judges but a working strength of 21,027, leaving 4,867 vacancies.
But behind these headline numbers is another, less visible layer to the problem: what happens to a case between filing, hearing, orders, and execution.
Legal information can remain scattered across files and formats. Orders may contain multiple directions and deadlines. Documents have to be located, interpreted, and acted upon by different people at different stages. A missed date or delayed action can push a matter further down the line.
It is this administrative layer of the justice system that Jupitice Justice Technologies is attempting to address.
“Backlog does not exist only because there are too many cases. It builds up because legal information sits fragmented across files and formats, actions wait in the wrong queue, deadlines get missed, and disputes end up in court before anyone has tried a resolution path that could have worked earlier. That is the layer we are trying to fix at Jupitice,” said Mansi Omar, Co-Founder and Chief Strategy Officer at Jupitice Justice Technologies.
The company’s approach is to bring disparate case information into a single structured record, allowing judges, registry staff, advocates, and case teams to work from the same information rather than repeatedly reconstructing a case history.
The technology layer also uses AI-assisted tools to extract information such as order dates, directives, and compliance deadlines, as well as to identify patterns across similar matters. But Jupitice says the objective is not to hand over legal decisions to machines.
“The review and the decision stay with the person, not the system,” Omar said.
That distinction is important as courts and legal institutions increasingly experiment with artificial intelligence. India’s own e-Courts programme is moving in this direction: the government says Phase III, backed by around Rs 7,210 crore, envisages digital and paperless courts, digitisation of case records, and the use of AI and optical character recognition for case analysis and forecasting.
The scale of the existing digital transformation is already substantial. According to the Department of Justice, about 660.36 crore pages of court records have been digitised, while more than 1.07 crore cases have been filed through e-filing. The National Service and Tracking of Electronic Processes system has processed 7.26 crore e-processes, of which 2.09 crore have been successfully delivered.
Jupitice is therefore positioning its proposition around what happens after information becomes digital: how that information is structured, verified, and moved through the workflow.
The company is building what Omar describes as the trust infrastructure around these tools. Its workflow includes audit logs recording who accessed or changed information and when, while source documents remain attached to AI-assisted extractions so users can verify what the system has identified. Quality checks are also built into the process rather than left to a final review.
“Without that layer, faster processing means very little,” Omar said.
The larger ambition, however, is to reduce the number of disputes that reach the courts in the first place.
Jupitice is developing digital channels for negotiation, mediation, arbitration, and Lok Adalat-style resolution. The idea is straightforward: if a dispute can be settled before litigation, it does not have to become another case competing for judicial time.
This shifts the conversation from merely disposing of pending cases to preventing avoidable cases from entering the system.
The NJDG data illustrates why that matters. The latest dashboard shows that nearly 29.2 lakh cases were instituted in the preceding month, while about 23.9 lakh were disposed of, indicating the continuing pressure created by fresh filings even as courts dispose of large numbers of cases.
For courts already carrying enormous caseloads, even seemingly small administrative delays can have a cumulative impact. A missed deadline can lead to another hearing. A document that cannot be located can result in rework. An order that is not acted upon promptly can prolong a matter even after the judicial decision has been made.
“None of this is about one clever feature. It is about connecting structured data, workflow, and human oversight so that fewer things fall through administrative gaps,” Omar said.
The intended gains, she said, are operational: fewer missed hearings, faster movement from an order to the action required on it, and less rework caused by information that could not be located in time.
That makes Jupitice’s proposition less about replacing the courtroom and more about rebuilding the processes around it.
Technology cannot create more judges or eliminate the flood of litigation. Nor can digitisation by itself solve a backlog running into crores of cases. But if information moves faster, deadlines are visible, administrative bottlenecks are reduced, and suitable disputes are diverted towards settlement, it could address one of the less-discussed contributors to India’s judicial backlog.
