Authenticity, Integrity and Admissibility of Digital Evidence in the Age of AI-Driven Cybercrime: Challenges for the Indian Legal System

Paper Code: AIJACLAV27RP2025

Category: Research Paper

Date of Publication: Feb 10, 2026

Citation: Ms. Mamta, “Authenticity, Integrity and Admissibility of Digital Evidence in the Age of AI-Driven Cybercrime: Challenges for the Indian Legal System”, 5, AIJACLA, 291, 291-300 (2025), <https://www.aequivic.in/post/authenticity-integrity-and-admissibility-of-digital-evidence-in-the-age-of-ai-driven-cybercrime-ch>

Author Details: Ms. Mamta, LLM Student

Download PDF

Abstract

The rapid development of artificial intelligence not only changed the nature of the phenomenon of cyber-crime, but also gave important burdens on the modern legal systems in terms of the evidence. In India, the growing popularity of deepfakes, ransomware attacks, automated intrusion-detection technologies, and identity-spoofing technologies make digital evidence less and less authenticable, preservable, and presentable in court. The available legislative means such as the Indian Evidence Act Section 65B was designed to handle traditional electronic records and cannot effectively deal with the instability, manipulation, and artificial creation of modern data. The judicial evolution in the Anvar P.V. to Arjun Panditrao cases, procedural necessities have been established, but the law doctrine remains many times behind the technological capability.

This paper is placed at the intersection between law and technology where the challenges of authenticity, chain of custody, integrity, and admissibility of digital evidence in AI-enabled environments are critically evaluated. Drawing upon comparative insights from the EU, UK, and United States, it argues for adaptive reforms such as AI‑assisted authentication, forensic audit trails, and verifiable metadata systems. This paper concludes that the reinforcement of the evidentiary standards is necessary to protect the due process, judicial certitude, and the confidence of the people in the digital justice system in India.

Keywords:- Digital Evidence, Artificial Intelligence, Admissibility of Evidence, Cybercrime and Indian Evidence Act.

Paper Code: AIJACLAV27RP2025

1.       Introduction

The recent progress in artificial intelligence (AI) has radically transformed the sphere of cybercrime and digital evidence. Deepfakes, voice synthesis, automated intrusions tools, and even AI-generated content technologies have not only elevated the levels of cyber offenses but also made the process of gathering, validating and presenting digital evidence in the court more difficult. Digital evidence is now the main point of criminal and civil processes, although its validity is being challenged more often in the age of algorithms and their manipulations.[1]

In India, the admissibility of electronic records is primarily governed by the Indian Evidence Act, 1872, particularly Section 65B, which was introduced through the Information Technology Act, 2000. The clause was also meant to guarantee the integrity and authenticity of electronic evidence through the requirement of certifications. Nevertheless, the body of legislation was written when electronic evidence was mostly in the form of fixed computer outputs, emails or server logs. The proliferation of AI-generated and AI-modified information has revealed the extreme weaknesses of this framework as such information may be created, modified, and reproduced with the minimal human involvement and without leaving any recognizable footprint.[2]

Judicial interpretation has tried to solve the procedural uncertainty by issuing landmark decisions like in Anvar P.V. v. P.K. Basheer[3] and Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal,[4] that made the nature of Section 65B certification mandatory.  Nonetheless, as much as these decisions have supported the ability to adhere to the procedures, it has not sufficiently addressed the underlying technological issues associated with authenticity, chain of custody, and forensic reliability in situations associated with the AI-driven manipulation. This means that there is a growing challenge on whether to accept evidence whose origin, integrity and originality can hardly be determined with recourse to ordinary legal tests.

The fact that there are no specialized forensic standards, judicial training in AI-based evidence or statutory appreciation of new authentication technologies further compounds the challenge. By contrast, other jurisdictions and particularly those in the European Union, the United Kingdom and the United States have started incorporating within their legal systems a forensic framework, audit trails, and risk-based approach to digital evidence.[5]

It is on this basis that the paper critically evaluates the sufficiency of the Indian evidence framework in combating the issues of AI-driven cybercrime. It claims that the admissibility regime in the Indian law will be destined to destroy due process and civic trust in the digital justice system without specific legal reform, integration of technology, and the capacity to build institutions.

2.       Research gap 

The current legal literature on digital evidence in India is mainly concerned with the procedural satisfaction in Section 65B of Indian Evidence Act. Nonetheless, little scholarly attention has been directed to the effect of artificial intelligence on the character and validity of digital evidence. Deepfake problems, manipulation of algorithms, and automated cybercrimes are also still not extensively researched in the Indian law. In addition, there is no comparative analysis on how other jurisdictions are addressing the same dilemma. This paper aims to address this gap by examining the lack of sufficiency of the current evidentiary framework as applied to the emerging AI-driven digital evidence and solutions to correct it.

3.       Research questions 

(i)   Whether the current legal system with references to the Indian Evidence Act, 1872 is sufficient to consider the problems of authenticity, integrity, and admissibility of digital evidence in a case related to AI-guided cybercrime?

(ii)   What is the effect of the development of artificial intelligence on the conventional principles of evidence and judicial norms of handling digital evidence in India?

(iii)   What legal and institutional changes would be required to successfully control and assess digital evidence in the era of artificial intelligence?

4.       Research objectives 

(i)       To examine the adequacy of the Indian legal framework, particularly the Indian Evidence Act, 1872, in dealing with issues of authenticity, integrity, and admissibility of AI‑generated or AI‑manipulated digital evidence.

(ii)     To examine the implications of artificial intelligence on the old principles of evidence and to measure the reaction of the judiciary to the technological intricate digital evidence. 

(iii)   To find and suggest appropriate legal and institutional changes, basing on the comparative international practices, to reinforce the evidentiary structure affecting the digital evidence in India. 

5.       Research methodology

The methodology of the given paper is a doctrinal and analytical research approach. The research relies on the analysis of the statutory provisions of Indian Evidence Act, 1872, and the judicial pronouncements on the relevance of electronic evidence. The comparative method has also been employed in the study of the legal developments in the other jurisdictions like the European Union, the United Kingdom and the United States. The secondary sources such as scholarly articles, reports and policy documents have been utilized as supporting materials to the analysis and come up with the reform-oriented conclusions. 

6.       The Digital Evidence in the AI Er

6.1   Digital Evidence: Meaning and Nature

Digital evidence is defined as either any information that can be relied upon in legal proceedings and has a probative nature, and is stored, transmitted, or generated in a digital form (Digital Evidence, 2011). Unlike its physical counterpart, digital evidence is intangible, volatile (easy to destroy), and relies on technological systems to create and maintain it (Digital Evidence, 2011). It can be easily modified, copied or removed and in most cases without leaving any traces, and this brings about a lot of concern as far as its reliability and authenticity is concerned.[6] 

The digital evidence should be differentiated with other similar concepts. The general concept of electronic evidence is data that has been stored or transmitted by using electronic devices like emails, call records, or CCTV footage.[7] Digital evidence is specifically defined as data that is presented in the binary format which has been handled by computer systems. AI‑generated evidence, however, represents a distinct category of data created or manipulated through artificial intelligence systems, including deepfakes, synthetic audio, and automated documents.[8]

In contrast to more traditional electronic records, evidence created by AI does not necessarily have a definite human behind it and can be created in unclear algorithmic ways. This complicates the process of proving authorship, originality, and integrity, which is the main conditions of admissibility provided by the Indian Evidence Act, 1872, especially the Section 65B.[9] 

6.2 AI and The Revolution of Cybercrime

The concept of artificial intelligence has greatly changed the nature of cybercrime because it allows more advanced types of manipulations of digitized objects. Deepfakes are fake audio and video generated by deep learning models, which can be classified as a grave threat of impersonation and misinformation. On the same note, voice-cloning technology has been abused in cases of financial fraud and identity theft and verification process in the old method is not dependable.[10]

Artificial intelligence has not merely changed the technologies of cyber-attacks but also made the processes of automated hacking and phishing generally efficient, allowing to carry out mass, accurately targeted cyber offenses with virtually minimal human organization. Further, the prosperity of AI-generated texts and fake personalities has undermined the boundary between genuine and fabricated online records, so it makes forensic authentication significantly more challenging.[11]

These dynamic changes make digital evidence increasingly difficult to prove. Traditional safeguards, such as metadata scrutiny or certification mandated by Section 65B, typically prove insufficient in unveiling algorithmic manipulation. The court system, in its turn, struggles with the lack of technical expertise required to evaluate AI-driven evidence and, as a result, increases the risk of mistaken dependence on created or altered data.

In turn, AI has radically changed the situation with evidence as it is now forced to reevaluate both current legal norms rigorously. Lacking the incorporation of modern technologies, the latest forensic procedures, and the increased judicial capacity- building programmes, the reliability of digital evidence, and, consequently, the fairness of the justice system, remains significantly compromised.[12]

7.       Legal System that Regulates the Use of Digital Evidence in India

7.1 Indian Evidence Act, 1872 

The third act, the Information Technology Act, 2000 has amended the Indian Evidence Act, 1872 that mainly governs the admissibility of digital evidence in India. The Act attempts to bring electronic records under the traditional evidentiary model, but the concept was developed prior to the emergence of the intricate consequences of artificial intelligence. 

7.1.1 Section 65B : Certificate Requirement

Section 65B provides the preconditions for making electronic records admissible, requiring that a certificate describing how the record was produced, which device was used, and whether the data is sound be furnished. The provision aims at ensuring authenticity and reliability by making the certificate a substitute of primary evidence.[13] However, the clause assumes a solid and recognizable source of information, which is becoming less possible due to AI-generated or AI-edited content. 

7.1.2 Section 45A : Expert Opinion 

Section 45A allows courts to accept the view of an investigator of electronic evidence. Although this amendment does not forbid technical assessment, it does not set standards on how AI-generated content or algorithmic procedures should be assessed. The lack of legal provisions that regulate the examination of complex AI systems limits the persuasive nature of expert testimony in cases that involve deepfakes, automated data creation, or algorithmic modifications.[14]

7.1.3 Section 22A : Oral Admissions

Section 22A says that oral acknowledgments about the content of electronic records are inapplicable except where the legitimacy of these records is at issue. This clause strengthens the principle of authenticity but provides very little on how the courts ought to determine authenticity in instances where digital evidence is either produced or modified by artificial intelligence. 

7.2 Interpretation of Digital Evidence by the Courts

7.2.1 Anvar P.V. v. P.K. Basheer, (2014)[15]

Anvar, P. V. v. P.K. Basheer, the Supreme Court declared that an electronic evidence can only be admissible in the event that it is in conformity with the Section 65B. The Court discarded the previous tradition of admitting electronic records as secondary evidence and emphasized the need of procedural protections to guarantee authenticity. It was an example of an extreme, technical position on digital evidence. 

7.2.2 Shafhi Mohammad v. State of Himachal Pradesh, (2018)[16] 

The Court in Shafhi Mohammad mitigated against the obligatory character of Section 65B, by that the certificate condition can be softened under circumstances where the party did not assume control over the electronic device.[17] Though hailed as a pragmatic response, this adjudication engendered legal uncertainty and inconsistency in the application of evidentiary standards.

7.2.3 Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal, (2020)[18]

The Supreme Court in Arjun Panditrao inverted Shafhi Mohammad and restated the Supreme Court position that the certification in Section 65B was mandatory. The Court made it clear that procedural compliance is central to the creation of authenticity and reliability of electronic evidence.[19] Nevertheless, the ruling was silent regarding the issues brought about by AI-produced evidence or automated data systems. 

7.2.4 Sonu @ Amar v. State of Haryana, (2017)[20]

In this case the Hon’ble Court once again reiterated that the objections on the mode of proof should be subjected in the trial stage and this strengthens the procedural inflexibility characteristic of electronic evidence.[21] This strategy stresses on technical compliance as opposed to substantive reliability. 

7.3 Major weaknesses of the current Framework 

The current legal framework has three major flaws, even though it has been elaborated in the courts: 

I.          Procedural Rigidity: The absolute demand of Section 65B certification often causes the omission of relevant evidence, including where authenticity is not actually at issue. 

II.        Lack of AI-Specific Recognition: Evidence generated or manipulated by AI is not included in the Evidence Act or judicial precedents, leaving the courts without directions on the issue of its possible bias in algorithms, independent creation, or detection of deepfakes. 

III.     Absence of Authentication Standards: There are no statutory or judicially accepted legitimacy of authenticating, integrity or provenance of AI-generated digital evidence, resulting in judicial ambiguity and disproportionate results. 

Thus, despite the fact that procedural clarity with regard to electronic evidence has been developed in Indian courts, the current framework is not well-prepared to face the technological aspects of AI-based cybercrime. 

8.    Objections to Authenticity, Integrity, and Admissibility of Digital Evidence in the AI Era 

The increasingly expanding dependency on artificial intelligence when it comes to producing and handling digital information introduces significant obstacles to outdated notions of evidentiary rule. Though compliance with the procedures is the focus of the Indian law, AI-based evidence raises more fundamental issues of authenticity, integrity, and the reliability of the judiciary that are not completely addressed yet. 

8.1 Challenges to Authenticity 

The authenticity requires indisputable evidence that evidence is real and has been provided by a credible source. This has placed this requirement on AI-generated content exponentially more onerous to satisfy. Deepfakes, synthetic media, voice cloning, and other technologies are used to create digital content of a very high level of realism, yet of an artificial nature, which is often impossible to distinguish without resemblance to genuine content.[22]

The old indicators that were used in the past such as metadata, hash values, device identification are insufficient in situations related to algorithmic manipulation. The content created by AI can have no identifiable human creator, making it far too complicated to assign and verify it.[23] The Indian Evidence Act, 1872, is silent in establishing authenticity when the material is created or modified through machine learning systems, thus creating a huge evidentiary gap. 

The lack of legal criteria to evaluate AI-created text increases the likelihood of admissible false evidence thus undermining the trustworthiness of judicial outcomes.

8.2 Challenges to Integrity and Chain of Custody

Integrity is the guarantee that the evidence has not been modified since its creation until the time when it is presented in court. The nature of digital evidence makes it prone to manipulation, copying and erasing. This problem is amplified in AI‑driven environments where data may be continuously processed, enhanced, or transformed by automated systems.

In section 65B of the Indian evidence Act, attention is given to certification and not forensic continuity. Although the provision offers the compliance of procedures, it does not touch on whether the data has been manipulated by algorithmic processes, automated decision-making, and training models.[24] Without an audit trail that can be verified or with standardized forensic procedures, it is virtually impossible to prove a continuous chain of custody.

It has been observed by scholars that traditional forensic procedures are inadequately equipped to evaluate AI-driven evidence because these systems can be considered black boxes with very little social visibility of data processing algorithms. This undermines the value of the evidentiary purpose of digital records and creates a greater margin of manipulation.

8.3 Objections to Admissibility and judicial Evaluation 

Indian courts are still keen on compliance procedure set out by Section 65B in order to ascertain admissibility. In the case, Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal revised the requirement of certification and stressed the form rather than the technology content.[25] While this ensures uniformity, it does not address the substantive reliability of AI‑generated evidence.

There is also a major restriction of courts to evaluate complex algorithmic evidence because of the absence of technical expertise and judicial guidelines. Reliance to expert testimony that is allowed under Section 45A is frequently ineffective since experts can disagree or have no access to proprietary AI systems.[26] This causes an evidence asymmetry and brings into question the freedom of due process and the right to trial.

In contrast, jurisdictions such as the United Kingdom and the European Union have begun developing forensic standards and regulatory frameworks to assess the reliability of AI‑generated evidence.[27] The Indian system, in its turn, is still working in the paradigm that was not meant to develop algorithmic decision-making and automated generation of data.

8.4 Requirement of a Recalibrated Evidentiary Approach

The difficulty of AI-driven evidence proves the drawbacks of a purely procedural method of admissibility. There is an urgent need for legal recognition of AI‑specific risks, development of forensic standards for authentication, and judicial training in emerging technologies. In the absence of such reforms, the current evidentiary regime may be insufficient in the future to deliver justice in a fair, accurate and trustworthy manner to the citizens of the digital justice system.

9.       Comparative Analysis: Digital Evidence in the AI Era Across International Relations

Comparative viewpoint shows that various jurisdictions have already started to modify their laws to suit the apparent legal difficulties of artificial intelligence, albeit with differing success. The jurisdictions like the European Union, the United Kingdom and the United States have followed a more technology sensitive approach in comparison with India which has not, as the legal framework has been largely procedural.

9.1 European Union 

EU has initiated a pro-regulatory stance on AI and digital evidence. The suggested Artificial Intelligence Act, 2021 categorizes AI systems as low or high risk and makes high-risk AI applications, such as law enforcement and judicial decision-making, highly responsible.[28] The EU system is based on the focus of transparency, traceability, and the supervision of people, without which the admissibility of AI-generated evidence is directly affected.

In addition, the eIDAS Regulation and the General Data Protection Regulation (GDPR) enhance the reliability of the evidence by requiring data integrity, the level of authenticity, and accountability in the digital process.[29] These tools make sure that electronic evidence can be covered by verifiable audit trails and traceable sources of data, decreasing a chance of manipulation.

In contrast to procedural certification that is so much strict in India, the EU model embraces substantive reliability, which concerns the process of data creation, processing, and storage.

9.2 United Kingdom 

The UK has established a relatively developed system of digital evidence forensics. The Code of Practice and Conduct of the UK Forensic Science Regulator provides specifications of the standards in the delivery of evidence, the digital forensics, and the validation of forensic tools.[30] These standards emphasise reproducibility, transparency, and documentation key elements for assessing AI‑generated evidence.

The evidentiary approach in the British courts is also flexible, which, according to the Criminal Practice Directions, permits the judges to determine the accuracy of the digital evidence with consideration to the expert evidence and forensic practices and not the strict adherence to the procedure.[31] This will allow courts to provide a more effective examination of algorithmic evidence where there is a case of automated systems or machine-generated data.

9.3 United States 

The United States has the Federal Rules of Evidence that regulate the admissibility of digital evidence, specifically Rule 901, that demands that the evidence be authenticated by means of reliable methods. The Daubert standard also requires that expert evidence should be both scientifically valid and empirically testable.[32]

The U.S. courts have been analyzing the accuracy of the algorithmic systems in more cases, particularly the ones that concern facial recognition, predictive policing, and automated decision-making. Judicial review is concerned with the error rates, transparency, and methodological reliability, and this guarantees the elevated standard of evidentiary credibility than the formalistic certification standards.[33]

9.4 Comparative Assessment 

Comparative analysis indicates that even though the main mode of compliance in India is still procedural compliance in Section 65B, there are other jurisdictions that have transited to substantive assessment of technological reliability. In the case of the EU, regulatory oversight is emphasised, the UK is concerned with forensic standards, and the U.S. is concerned with scientific validity and expert scrutiny.

The evidentiary system of India is wanting in: 

·   AI-specific authentication standards, 

·   Forensic audit mechanisms, and

·           Algorithmic evidence judicial guidelines.

Such relative difference suggests the critical necessity of India to re-tune its legal system to address the threats of AI-driven digital evidence.

10.   Way Forward and Recommendations

The growing sophistication of AI-based digital evidence requires a shift in the evidentiary paradigm of India towards a more technology-sensitive and reliability-focused paradigm. The reforms needed include the following to promote fairness, accuracy, and judicial consistency.

10.1           Statutory Reform of the Indian Evidence Act 

Section 65B of the Indian Evidence Acts requires a reconsideration to capture AI generated and AI processed evidence. The provision must be changed to: 

·   Identify AI-generated evidence as a separate category of evidence; 

·         Require reporting of method of generation, algorithmic engagement and sources of data;  and

·         Present assumptions on technological reliability as opposed to certification only.

This kind of reform would bring the Indian law in accord with the international best practice and decrease the overreliance on formalistic adherence.[34]

10.2           The forensic and authentication standards are developed

India needs to embrace homogeneous forensic standards of digital evidence especially of: 

·         Deepfake detection, 

·         Metadata verification, 

·         Algorithmic audit trails, and

·         Chain‑of‑custody documentation.

Based on the UK Forensic Science Regulator framework, the statutory guidelines can be presented in a form of digital forensic laboratories and expert witnesses to provide transparency and reproducibility of the evidence.[35]

10.3      Capacity Building of the Judiciary and the Expert Involvement 

There is a growing number of judges who are faced with technologically complex evidence that they lack sufficient training in. Periodic judicial education programmes on AI, digital forensics, and algorithmic decision‑making should be institutionalised.

Also, the courts need to be given the authority to introduce independent technical experts like the amici curiae in situations where there is a complicated AI-generated evidence, as was practiced in the federal courts of the U.S.[36]

10.4 Embracing a Risk-Based and Technological-Neutral Strategy

India should shift towards the risk-based regulatory system in which digital evidence is assessed in accordance with the possible effect on rights and liberties. This would be achieved by regulating through the EU Artificial Intelligence Act, which would keep pace with technology and avoid the need to amend legislation on a regular basis.[37]

This type of model will provide flexibility, accountability, and judicial adaptability during a period of the fast development of AI systems.

11.   Conclusion 

With the introduction of artificial intelligence, the digital evidence has essentially changed the nature of it, making traditional concepts of authenticity, integrity, and admissibility challenged. Although the Indian legal system has made the first steps towards researching electronic evidence with the help of the statutory provisions and judicial interpretation, it is still unprepared to meet the challenges posed by AI-introduced technologies.

Further, the maintenance of the traditional compliance within the frameworks of Section 65B in the absence of proper attention to the technological realities threatens to compromise the aspects of evidentiary reliability and due process. As a comparative analysis indicates, some jurisdictions like the European Union, the United Kingdom and the United States have assumed more subtle positions that lay emphasis on forensic reliability, transparency, and accountability. 

This paper concludes that India needs to understand that it should not be on a formalistic evidentiary model but embrace a technologically informed and future proofed model. Critical reforms in the legal system, institutional capacity building, and the incorporation of forensic standards are necessary to make sure that digital evidence is still a means of justice and not of ambiguity. Such recalibration is the only way to make the legal system responsive to the problems of AI-driven cybercrime and guarantee that people trust the digital justice system.

[1] Shreya Singhal v Union of India (2015) 5 SCC 1 (SC); OECD, Artificial Intelligence in Society (OECD Publishing 2019).

[2] Information Technology Act 2000, s 65B; Orin S Kerr, ‘Digital Evidence and the New Criminal Procedure’ (2005) 105 Columbia Law Review 279.

[3] Anvar P.V. v P.K. Basheer (2014) 10 SCC 473 (SC).

[4] Arjun Panditrao Khotkar v Kailash Kushanrao Gorantyal (2020) 7 SCC 1 (SC).

[5] European Commission, Proposal for a Regulation Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) COM (2021) 206 final.

[6] Kerr (n 4) 281–82.

[7] Kerr (n 4) 284.

[8] Indian Evidence Act 1872, s 65B.

[9] Anvar P.V. (n 5).

[10] European Commission, Tackling Deepfakes in the Digital Age COM (2021) 206 final.

[11] INTERPOL, Artificial Intelligence and Cybercrime: Challenges and Responses (2022).

[12] UK Forensic Science Regulator, Code of Practice and Conduct (2020); Federal Rules of Evidence 901 (US).

[13] Indian Evidence Act 1872, s 65B.

[14] Indian Evidence Act 1872, s 45A.

[15] Anvar P.V. (n 5).

[16] Shafhi Mohammad v State of Himachal Pradesh (2018) 2 SCC 801 (SC).

[17] ibid.

[18] Arjun Panditrao (n 4).

[19] ibid.

[20] Sonu @ Amar v State of Haryana (2017) 8 SCC 570 (SC).

[21] ibid.

[22] European Commission (n 10) 5.

[23] Kerr (n 4) 284.

[24] Frank Pasquale, The Black Box Society: The Secret Algorithms That Control Money and Information (Harvard University Press 2015) 3–5.

[25] Arjun Panditrao (n 4).

[26] Indian Evidence Act 1872, s 45A.

[27] UK Forensic Science Regulator, Code of Practice and Conduct (2020); European Commission, Artificial Intelligence Act, COM (2021) 206 final.

[28] European Commission, Proposal for a Regulation Laying Down Harmonized Rules on Artificial Intelligence (Artificial Intelligence Act), COM (2021) 206 final.

[29] Regulation (EU) No 910/2014 on electronic identification and trust services for electronic transactions in the internal market [2014] OJ L257/73; Regulation (EU) 2016/679 (General Data Protection Regulation) [2016] OJ L119/1.

[30] UK Forensic Science Regulator, Code of Practice and Conduct (2020).

[31] Criminal Practice Directions 2015 (UK).

[32] Federal Rules of Evidence 702 (US); Daubert v Merrell Dow Pharmaceuticals Inc 509 US 579 (1993).

[33] State v Loomis 881 NW 2d 749 (Wis 2016).

[34] Indian Evidence Act 1872, s 65B; Law Commission of India, Report No 185: Review of the Indian Evidence Act (2003).

[35] UK Forensic Science Regulator, Code of Practice and Conduct (2020).

[36] Federal Rules of Evidence 702 (US); Daubert v Merrell Dow Pharmaceuticals Inc 509 US 579 (1993).

[37] European Commission, Artificial Intelligence Act, COM (2021) 206 final.