The evolution of admissibility of digital evidence in India
Neerav Merchant
AQUILAW, Mumbai
neerav.merchant@aquilaw.com
Poonam Ashar
AQUILAW, Mumbai
poonam.ashar@aquilaw.com
Transition from paper trails to digital footprints has transformed the proving of facts in courts. E-mails, messaging app logs, CCTV archives, server logs, etc., routinely feature in civil and criminal litigations. The repealed Indian Evidence Act, 1872 (“IEA”) adapted to this reality through sections 65A-65B, primarily, drafted for an era of standalone computers, optical or magnetic media. Similarly, the Bharatiya Sakshya Adhiniyam, 2023 (“BSA”) seeks to consolidate and modernise evidentiary rules, including dealing with electronic and digital records.
This article examines three issues: (i) whether the BSA modernises admissibility or substantially replicates Section 65B of the IEA; (ii) whether BSA is technology-neutral to accommodate evolving platforms such as Artificial Intelligence (“AI”) systems and synthetic media, and (iii) whether BSA provides adequate safeguards against deepfakes, data manipulation, and opaque AI outputs.
Statutory architecture: from the IEA to the BSA
The IEA
Under the IEA, electronic records were governed by a dual structure: general secondary evidence rules in sections 63 and 65 and special provisions in sections 65A-65B. Section 63 defined secondary evidence, while Section 65 explained permissibility of its use where the original could not be produced. Inserted in 2000, Section 65B1 operated as a complete code,2 for electronic records and an exception to the general rules.
Section 65A served as an introductory provision, declaring the applicability of section 65B to evidence generated by a “computer” during its activities. Section 65B deemed information contained in an electronic record and produced by a computer during its ordinary course of activities to be a document, by satisfying conditions under section 65B (2).
Section 65B (4) mandated procurement of a certificate from a person responsible for the operation or management of the relevant computer system, identifying the record, device particulars, and manner of production, which was affirmed as a mandatory condition precedent by the Supreme Court in Anvar P.V. v. P.K. Basheer & Ors.,3 and reaffirmed in Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal.4
The BSA
The new code expressly recognises the centrality of electronic and digital records. Section 2(1)(d) includes “electronic and digital records”5 in the definition of “document”, while section 2(1)(e) recognises electronic information as a form of documentary evidence,6 placing digital records on greater parity with traditional documents. Section 2(2) of the BSA expressly provides for referring to the Information Technology Act, 2000 (“IT Act”) for the terms not defined therein.
Section 61 clarifies that an electronic or digital record cannot be denied admissibility merely because of its electronic form, subject to section 63. This is a clear shift from the earlier judicial concerns that treated digital evidence as inherently exceptional. Sections 62 and 63 broaden the earlier section 65B framework by extending it beyond optical and magnetic media to semiconductor memory and “any communication device”7 thus accommodating cloud-based storage, distributed systems, and virtualised environments.
Section 63(4) of the BSA read with the prescribed Schedule succeeds section 65B(4) and is now the central statutory mechanism for authenticating secondary electronic records. Under section 63(4), the requirement of “lawful control” is functionally preserved through BSA’s insistence that the person issuing the certificate has the operational responsibility to the relevant system, a feature that assumes particular significance in third party custody scenarios involving telecom providers, social media platforms, cloud service providers, and government databases.
The jurisprudence re WhatsApp chats is a good example of significance of certification, wherein it was affirmed that screenshots or printed chats, unsupported by a compliant certificate, are inadmissible.8 Section 66 separately addresses proof of electronic signatures.
The architecture is coherent, but its effectiveness depends on procedural compliance and forensic competence.
Expanded and technology-neutral definition of “electronic record”
A core feature of the BSA is its segregation from device specific terminology towards function based, technology-neutral categories. The BSA adopts broader terminologies such as electronic and digital records, computer resources, and communication devices. This purposive drafting seeks to future-proof evidentiary rules against technological change and addresses a limitation of the IEA, whose provisions were developed around localised computer systems and physical media.
Neither IEA nor BSA touches upon the evidentiary status of AI-generated outputs, predictive analytics, or algorithmic inferences, despite the growing reliance on such newer and AI lead technologies in commercial decision-making and investigative processes. For instance, India’s Ministry of Law and Justice confirmed that AI is being integrated into policing and law enforcement to enhance crime detection, surveillance, and criminal investigations.9
The BSA’s generality is double-edged. Its broad definitions may accommodate AI-generated material, but the absence of AI-specific safeguards such as transparency requirements, disclosure concerning training data, error-rate reporting or independent forensic certification creates uncertainty regarding reliability and admissibility.
Separation of admissibility and presumptions
One of the distinguishing features in the BSA is the clearer separation between (i) admissibility, (ii) presumptive authenticity, and (iii) evidentiary weight. Sections 61 and 63 address the modalities for admitting electronic records, while Sections 85 to 87 address presumptions and probative status of such records. Section 57 provides that an electronic or digital record produced from proper custody should be treated as primary evidence unless disputed, moving away from treating digital material as inherently ‘secondary’. In turn, Section 86 creates presumptions relating to electronic signatures and specified secure digital processes where prescribed technological and procedural safeguards exist.
Chain of custody and forensic capability
The reliability of digital evidence depends on the integrity of its chain of custody and forensic infrastructure available to capture, preserve, and analyse it. Existing gaps include inconsistent seizure protocols, limited standardisation of forensic imaging, inadequate training of police and court staff, and uneven access to up to date tools for log analysis, hash verification, and deepfake detection. The BSA’s certificate form and presumptions implicitly assume robust forensic processes. There is, however, no comprehensive statutory code yet governing digital chain of custody procedures or mandating minimum forensic standards across investigating agencies and forensic laboratories.
Other jurisdictions with useful examples include, the UK’s ACPO Good Practice Guide, EU E-Evidence regime, US Federal Rules of Evidence 901–902 and OLAF guidelines, addressing non-alteration, audit trails and competent access; cross-border evidence gathering; advanced authentication standards; and structured acquisition, imaging and chain-of-custody documentation.
The Standing Committee on Home Affairs (“Standing Committee”) flagged the absence of statutory safeguards to ensure a proper chain of custody for electronic and digital records obtained during investigations, observing that the absence of a formal chain-of-custody requirement leaves such records vulnerable to manipulation. Upon assessing these comparative jurisdictional practices, the Standing Committee recommended inserting a provision mandating the secure handling and processing of digital evidence, with its integrity ensured.
Implementation challenges under the BSA
Despite its coherent architecture, the implementation of the BSA faces several challenges. Firstly, extensive jurisprudence and courtroom practice have developed around Section 65B of the IEA. Hence, the transition to the BSA requires the judges and legal practitioners to learn its revised terminology. Secondly, varied jurisdictions and litigant categories in access to certification, hash verification, digital imaging, and expert testimony under the BSA, may result in uneven evidentiary quality, risking inconsistent application of certification requirements and may affect procedural fairness, particularly for resource-constrained litigants. Thirdly, the BSA operates within a fragmented regulatory ecosystem. Its evidentiary provisions intersect with parallel statutory regimes, including the Information Technology Act, 2000, the Digital Personal Data Protection Act, 2023, and criminal procedure law governing search, seizure, preservation, and interception. These frameworks were drafted at different intervals and have not yet been fully harmonised with the BSA’s terminology and conceptual categories. The absence of cross-statutory alignment may generate uncertainty re admissibility standards, data retention obligations, and lawful access thresholds.
Looking forward, three (3) reform vectors merit attention, (i) Doctrinal clarification on AI-generated evidence under Section 63 of the BSA, including standards for transparency and probabilistic outputs; (ii) Institutional capacity-building, including investment in digital forensic infrastructure, laboratory accreditation standards, uniform hash verification protocols, and sustained judicial training in technological evidence; and (iii) Cross-statutory harmonisation, ensuring coherence between evidentiary doctrine and data protection, surveillance, and platform governance regimes, thereby reducing the risk of unusable evidence and overbroad data collection practices.
Conclusion
The BSA has been evolutionary from IEA’s exception based treatment of electronic records to an integrated framework recognising electronic and digital records as central, not peripheral, to contemporary adjudication. By broadening definitions, formalising certificate requirements, and distinguishing admissibility from presumptions and weight, the debate moves beyond the threshold question “Is this electronic record admissible?” towards the more substantive inquiry “Is this digital evidence reliable, authentic, and fairly obtained?”.
The structural shift reflects an implicit acknowledgment that digital evidence is not an exception to evidentiary doctrine but a routine feature of modern litigation. The BSA’s technology-neutral approach shifts the paradigm from medium-specific formalism towards principles capable of adapting to changing technologies. Its success depends less on statutory drafting than on robust forensic practices, judicial vigilance, and investments in technology, infrastructure and training.
---
Notes
1 Inserted by Act 21 of 2000, S. 92 and Schedule II (w.e.f. 17-10-2000).
2 Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal & Ors., (2020) 7 SCC 1 at paras 36-37 & 60.
3 Anvar P.V. v. P.K. Basheer, (2014) 10 SCC 473.
4 Supra at 2.
5 BSA, S. 2(1)(d).
6 BSA, S. 2(1)(e).
7 BSA, S. 63.
8 Dell International Services India (P) Ltd. v. Adeel Feroze, 2024 SCC OnLine Del 4576, at Para 11.
9 Press release dated 25 February 2025 of the Indian Ministry of Law and Justice https://www.pib.gov.in/PressReleasePage.aspx?PRID=2106239®=3&lang=2