Competition Law and Layered Platform Dominance Theories

1. Introduction

Smart Meter Fraud Detection Systems are technological and regulatory mechanisms used by electricity suppliers and distribution network operators to identify electricity theft, meter tampering, bypassing, manipulation of consumption records, and other forms of dishonest energy abstraction. Unlike conventional meters, smart meters generate frequent consumption data and may record technical events such as power interruptions, communication failures, unusual voltage conditions, and suspected tampering. Fraud-detection systems analyse these data to identify patterns requiring investigation. However, an algorithmic alert should ordinarily be treated as an indicator for investigation rather than automatic proof of fraud.

2. Fraud Detection Technologies

Modern fraud detection combines smart-meter event logs, consumption-pattern analysis, anomaly detection, machine learning and physical inspection. Algorithms can compare a customer's historical consumption with later readings, comparable premises and expected load characteristics. Sudden unexplained reductions, repeated abnormal power-down events, meter-cover interference or significant differences between expected and recorded consumption may generate alerts.

Network-level analysis can also compare electricity entering a distribution area with aggregated metered consumption. Persistent unexplained differences may indicate technical losses, faulty meters or electricity theft. Therefore, fraud systems must distinguish deliberate interference from equipment malfunction, communication failure and legitimate changes in consumer behaviour.

3. Legal and Regulatory Framework

Electricity theft through meter manipulation is specifically addressed in jurisdictions such as India. Section 135 of the Electricity Act 2003 covers dishonest interference with meters and the use of tampered meters where the statutory requirements are satisfied. Judicial decisions emphasise that dishonesty and reliable evidence remain important even where sophisticated technological detection is used.

A useful comparative principle also appears in UK regulation. Ofgem describes meter tampering as interference causing a meter to record less energy than actually consumed or bypassing the meter altogether. It stresses that apparent signs of tampering can also result from damage or faults and therefore require investigation.

4. Case Law – Smt. Jagdish Narayan v North Delhi Power Ltd (2007)

Case Name/Citation: Smt. Jagdish Narayan v North Delhi Power Limited & Anr., Delhi High Court, 18 April 2007.

Facts: The dispute concerned allegations of dishonest abstraction of electricity based upon indications of meter interference and consumption-related evidence.

Legal Issue: Whether external signs of meter tampering and consumption patterns were sufficient to establish dishonest abstraction of energy.

Judgment: The Delhi High Court emphasised that theft requires the element of dishonesty. External symptoms of tampering cannot automatically establish dishonest abstraction without sufficient tangible evidence connecting the consumer to the manipulation.

Legal Principle/Ratio: Fraud detection indicators create grounds for investigation, but the legal conclusion of electricity theft requires persuasive evidence establishing dishonest interference.

Significance: The case is particularly relevant to AI-based smart-meter fraud detection. An anomaly score or unusual consumption profile should not, by itself, determine liability.

5. Case Law – BRPL v Rajesh Devi (2023)

Case Name/Citation: BRPL v Rajesh Devi, Ct. Case No. 205/2018, decided 7 August 2023.

Facts: An electricity meter was removed and tested, with the utility alleging illegal cut marks and substantial under-recording. Consumption analysis was also relied upon.

Legal Issue: Whether laboratory and consumption evidence sufficiently proved electricity theft through meter tampering.

Judgment: The court acquitted the accused. Importantly, the laboratory report central to the allegation had not been properly proved through the person responsible for testing the meter, and the consumption evidence did not sufficiently substantiate the alleged theft.

Legal Principle/Ratio: Technical fraud evidence must itself satisfy evidentiary requirements; merely producing an analytical conclusion does not necessarily establish dishonest abstraction.

Significance: Smart-meter fraud systems therefore require auditability, evidential integrity and human verification.

6. Case Law – BSES Rajdhani Power Ltd v Abdul Aziz (2024)

Case Name/Citation: BSES Rajdhani Power Ltd v Abdul Aziz & Anr., decided 31 July 2024.

Facts: Meter testing identified cut marks, burnt terminals and abnormal power-down events.

Legal Issue: Whether these technical abnormalities demonstrated dishonest abstraction.

Judgment: The court scrutinised whether abnormal events could actually be connected with dishonest abstraction, noting uncertainty about when particular power-down events occurred.

Legal Principle/Ratio: Technical abnormalities must be linked logically and evidentially to the alleged fraudulent conduct.

7. Regulatory Significance

Effective smart-meter fraud governance should therefore combine automated detection with physical inspection, secure event logs, reliable laboratory testing, documented chain of custody and an opportunity for consumers to challenge allegations. Fraud-detection models must also minimise false positives because unusual consumption may arise from legitimate behavioural changes or technical faults.

Conclusion

Smart Meter Fraud Detection Systems can significantly strengthen electricity-market integrity by identifying suspicious consumption and tampering patterns at scale. Nevertheless, algorithmic suspicion is different from legal proof. The case law demonstrates that regulators and utilities should combine digital analytics with corroborating physical and evidential material before imposing theft liability. This creates a balanced framework in which smart technology supports enforcement while preserving procedural fairness and evidentiary reliability.