INDEPENDENT POLICY & RESEARCHConceptual framework Edition 0.16.4
Rival nations & shared rivers · 1960 framework; disputed after April 2025

A shared river needs a way to handle disagreement.

Could a mutually authorized, auditable information process make a water disagreement easier to examine while protecting each country’s decision rights?

Case 03 of 14 · Documented event · Proposed MAR application
The practical opportunity

Help authorized water officials reconcile approved observations and technical questions before a data disagreement becomes a wider political confrontation.

A long-standing water-information and disagreement framework between rivals, with its political rupture made explicit.

What happened

India and Pakistan signed the Indus Waters Treaty in 1960 with World Bank assistance. The Bank’s account describes a Permanent Indus Commission with a commissioner from each country and procedures for handling questions, differences, and disputes. World Bank · Updated 21 June 2023

The World Bank describes its own role as limited and procedural. The treaty’s history includes both cooperation through periods of tension and disagreements requiring separate institutional processes. World Bank · Updated 21 June 2023

India announced on 23 April 2025 that it would hold the treaty in abeyance. India Ministry of External Affairs · 23 April 2025

Pakistan rejected India’s announcement in its 25 April briefing. This case examines the historical cooperation framework; it does not assume uninterrupted operation or resolve the contested legal positions. Pakistan Ministry of Foreign Affairs · 25 April 2025

The coordination question

A shared technical record and a recognized disagreement process can give rivals alternatives to unilateral accusation. The 2025 rupture also demonstrates that a long record of cooperation does not make a framework immune to a security crisis.

What cooperation actually took place

The countries established a recurring bilateral mechanism for water information and technical questions, alongside distinct routes for disagreements. The World Bank’s account shows both the institutional achievement and its limits. Pakistan’s April 2025 response makes clear that India’s announced abeyance was contested.

The political divide

Rival neighbours; contested framework

India and Pakistan have an enduring adversarial relationship. The historical treaty architecture is relevant precisely because cooperation and serious conflict coexisted; its interruption must remain part of the lesson.

How could MAR benefit both sides?

Potential gains in a mutually authorized civilian MAR process.

India

An agreed MAR process could help Indian officials document approved observations, explain a technical position, and obtain a clear record of the questions requiring a counterpart’s response.

Pakistan

Pakistani officials could receive clearer approved information, challenge discrepancies with a traceable record, and distinguish an unanswered question from a confirmed departure from an agreement.

Each participant would need to judge the value, disclosure burden, and effect on its own civilian responsibilities. A benefit for one side does not by itself establish a mutually useful arrangement.

How could this lessen the chance of war?

The following pathway is Safe Nations’ proposal, rather than a measured historical result.

  1. Reduce avoidable factual disputes. Comparable, time-stamped records can narrow disagreement about what was observed before participants debate its meaning.
  2. Offer a recognized route to challenge. Each country can question a report through agreed institutions while preserving its position on the underlying dispute.
  3. Keep civilian needs visible. An approved account of water-service needs can help leaders assess the human consequences of proposed choices.

What would limit that effect?

Any exercise would require fresh, explicit authorization from the relevant parties. AI cannot restore a treaty mechanism, adjudicate water rights, settle the abeyance dispute, or guarantee that a technical finding changes political decisions.

What could we examine first?

Measure time to resolve documentary discrepancies, the share of claims supported by an agreed record, unresolved questions, and burdens on each party. A fair result must show improvement for both participants without concealing a legitimate disagreement.

Examine the evidence standard

Who would cooperate?

The roles below describe a proposed exercise based on the event. They are not a record of MAR participation.

Participant 01

Authorized water authorities from both countries

Agree on the exercise, permitted records, and the civilian question to examine.

Participant 02

Bilateral technical reviewers

Compare observations and explain discrepancies while retaining competing interpretations where necessary.

Participant 03

Competent diplomatic and dispute institutions

Determine applicable authority and handle issues beyond a technical information exercise.

A shared, verified information picture

Participants exchange only approved information, can challenge a report, and remain responsible for their own decisions.

How a MAR-style response could work

  1. Secure a common mandate. Define one civilian information task and obtain each side’s permission before any sharing.
  2. Reconcile approved records. Compare observation times, units, revisions, and missing information without inventing measurements.
  3. Document unresolved questions. Give both sides equal ability to challenge a finding and preserve their respective explanations.
  4. Refer and review. Use the authorized human process for unresolved matters and review whether the record improved civilian planning.

What AI could contribute

AI could organize permitted documents, compare units and versions, and identify missing administrative information. Any technical or legal interpretation would require competent human review.

Who would retain authority

Governments and the institutions they lawfully authorize would retain their responsibilities. No generated recommendation would determine water allocations or direct infrastructure operations.

Review the proposed controls

Who could benefit—and how?

These are possible benefits, not measured historical results.

Water officials in both countries

A clearer common record could reduce repeated administrative disagreement.

How to test the benefit
Document conflicts, correction time, and unresolved requests, reported separately for each side.

Farmers and essential-service communities

Better-supported civilian planning could reduce uncertainty where usable information and physical resources are available.

How to test the benefit
Planning revisions caused by avoidable information errors and unmet essential needs.

Public oversight bodies

Traceable claims could make the basis of a decision easier to scrutinize.

How to test the benefit
Whether an independent reviewer can distinguish observations, assumptions, and disputed interpretations.

Limits that matter

This is a historical institutional case with an explicit political rupture. Better information cannot create water, erase unequal dependencies, or substitute for a mutually accepted legal and diplomatic process.

A fair test of the added value

Use historical public or synthetic documents in an offline exercise approved by the participating institutions. Compare established document review, an improved non-AI process, and reviewed AI assistance, with equal access to evidence.

What would challenge the proposal?

Reject a result that invents consent, presents a disputed interpretation as settled, obscures unequal burdens, or turns a missing observation into an allegation of hostile intent.

Use authorized historical or synthetic records. Where the historical record lacks a necessary detail, label the exercise assumption rather than presenting it as a fact about the incident.

How uncertainty should be handled

Related MAR hypotheses

These source connections suggest questions to investigate. Their status remains untested hypotheses.

Inspect hypotheses in the Resilience Lab

Sources and interpretation

Primary institutional sources for the historical record. The workflow and benefit analysis above are Safe Nations’ proposals. Reviewed 9 November 2026.