Can A Fake Video Send Someone To Jail? India's Deepfake Evidence Crisis in India

Home › Blog › Can A Fake Video Send Someone To Jail? Evidence Law × Cyber Law × AI Can A Fake Video Send Someone To Jail? Inside India's Deepfake Evidence Crisis — Laws, Punishments, Landmark Cases & Your Rights By Aditya Kuwad (CA) & Siddhant Kuwad (LL.M.)  |  Published: May 28, 2026  |  Last Updated: May 28, 2026  |  ~9,500 words  |  35 min read Table of Contents Opening Scene: A Fabricated Confession Direct Answer — Can A Fake Video Send Someone To Jail? What Is a Fake Video / Deepfake? Why This Matters in India Can Fake Videos Become Evidence in Court? How Indian Courts Examine Digital Evidence What Laws Apply — The Complete Framework What If Someone Submits Fake Evidence? What If Someone Creates or Distributes a Deepfake? Landmark Cases & Real-World Examples How Police, Lawyers & Courts Verify Videos What Citizens Should Do The Future of Evidence in India Key Takeaways Frequently Asked Questions 1. Opening Scene: A Fabricated Confession Imagine this. A man is arrested in a tier-two Indian city. The prosecution presents a video in court — grainy CCTV footage, apparently showing the accused confessing to a crime outside a shop. The video is played. The magistrate watches. The accused protests — that is not me, that never happened . But the footage looks real. The face matches. The voice sounds right. Three months later, a forensic examiner reveals the truth: the video was a deepfake. The face had been digitally transplanted onto a different person's body. The voice was cloned using publicly available audio samples from the man's social media. The confession never took place. But for ninety days, the man sat in a jail cell because of a video that never depicted reality. This is not a hypothetical future. This is the crisis Indian courts are navigating right now. In the era of generative AI, where a teenager with a laptop can fabricate photorealistic video evidence in under an hour, the very foundation of what courts call "the best evidence" — seeing and hearing — is crumbling. The question India must answer is both urgent and terrifying: Can a fabricated video actually put an innocent person behind bars? 2. Direct Answer — Can A Fake Video Send Someone To Jail? The Short Answer Yes, it is legally possible — and the risk is growing. If a fake or deepfake video is submitted as evidence in an Indian court and is not detected as fabricated, a judge or magistrate could rely on it to convict an accused person. Indian evidence law treats electronic records — including videos — as admissible evidence under Sections 61–65 of the Bharatiya Sakshya Adhiniyam (BSA), 2023 , provided they satisfy certain procedural conditions including a mandatory certificate under Section 63(4) BSA . However, the legal system provides safeguards. If the video is later proven to be fabricated, the conviction can be overturned, and the person who submitted the fake evidence faces up to 7 years imprisonment under Section 229 of the Bharatiya Nyaya Sanhita (BNS), 2023 . The creator of the deepfake itself faces prosecution under IT Act Sections 66C, 66D, and 66E (up to 3 years each), BNS Section 336 (forgery, up to 7 years), and the IT Rules 2026 which mandate 3-hour takedowns of flagged deepfakes. The real danger lies in the gap between submission and detection — the period during which a court may trust a fake video because the technology to fabricate it has outpaced the technology (and training) to detect it. 3. What Is a Fake Video / Deepfake? Can a fake video be created using AI? Yes — and the technology has become alarmingly accessible. A "deepfake" is a piece of synthetic media — a video, image, or audio clip — in which a person's likeness, voice, or actions are digitally fabricated or altered using artificial intelligence, typically deep-learning neural networks, to create content that appears authentic but depicts events that never occurred. The term combines "deep learning" and "fake." Legal Definition — IT Rules 2026 Synthetically Generated Information (SGI) Under Rule 2(1)(wa) of the IT (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2026 — notified on February 10, 2026 and effective from February 20, 2026 — synthetically generated information means information that is "artificially or algorithmically created, generated, modified or altered using a computer resource, in a manner that appears reasonably authentic or true" and is "capable of being mistaken for an actual person or real-world event." Deepfakes are not limited to videos of celebrities. The technology is now used for face-swapping (placing one person's face onto another's body), voice cloning (replicating someone's voice from a few seconds of sample audio), lip-sync manipulation (altering a speaker's lip movements to match fabricated dialogue), and full synthetic generation (creating entirely fictional people who appear real). According to the Identity Fraud Report of 2025 , deepfake attempts occurred at a rate of one every five minutes globally in 2024, and deepfakes accounted for 40% of all biometric fraud. Between 2022 and 2023, the number of deepfakes increased by 3,000%, with volumes peaking in mid-2023. For the Indian legal system, this means that video and audio evidence — once considered among the most persuasive forms of proof — can no longer be taken at face value. The courtroom assumption that "seeing is believing" has been permanently destabilised. 4. Why This Matters in India Is deepfake a serious threat in India? Absolutely — and the scale is staggering. India is the world's second-largest internet market, with over 900 million internet users. The country has one of the highest rates of social media penetration globally, with WhatsApp alone serving over 500 million Indian users. This creates a vast surface area for deepfake distribution. The India Cyber Threat Report, 2025 by the Data Security Council of India labelled the severity of deepfake-enabled cybercrime as "critical" for the upcoming years, predicting that deepfake technology would be used to create compelling malicious content including fake video and audio messages from trusted sources. India's judiciary, meanwhile, handles over 40 million pending cases. Magistrates and judges — many of whom received their legal training before the smartphone era — are now being asked to assess the authenticity of AI-generated video evidence without standardised forensic tools or uniform training protocols. The Delhi High Court itself took judicial notice of this crisis when, in Nirmaan Malhotra v. Tushita Kaul (2024 SCC OnLine Del 4326), it observed that courts now operate in "an era of deepfakes" where the burden of proving the authenticity of photographic evidence rests firmly on the party presenting it. "The rights of a plaintiff cannot be rendered otiose in this world of rapidly developing technology." — Justice Saurabh Banerjee, Delhi High Court, in Sadhguru Jagadish Vasudev v. Igor Isakov , CS(COMM) 578/2025 (May 30, 2025) The crisis is not merely theoretical. In the lead-up to the 2024 general elections, industry reporting estimated over 50 million AI-generated voice-clone calls circulated across regional languages. Deepfake videos of politicians making inflammatory statements spread on WhatsApp before fact-checkers could respond. The Election Commission urged political parties to exercise restraint, but the viral nature of messaging platforms made containment nearly impossible. This political dimension pushed the Central Government toward the legislative action that eventually became the IT Rules 2026. Facing a Cyber Law or Evidence Issue? The EduLaw provides expert legal consultations on digital evidence, cyber crimes, deepfake litigation, and court-related matters. Speak with qualified professionals who understand both the law and the technology. Explore Our Legal Services → 5. Can Fake Videos Become Evidence in Court? Can a fake video be used in court as evidence? Under Indian law, yes — any video, including a fake one, can be submitted as evidence. The question is whether it will be admitted and whether the court will rely on it. These are three distinct legal thresholds, and understanding the difference is critical. All electronic records, including video recordings, are governed by Sections 61 to 65 of the Bharatiya Sakshya Adhiniyam (BSA), 2023 , which replaced the corresponding provisions of the Indian Evidence Act, 1872. Section 57 of BSA defines "electronic record" expansively, and Section 63 establishes the conditions under which electronic records can be admitted as secondary evidence. This means a video does not need to be produced on its original recording device; a copy is admissible provided a valid certificate is furnished. The gateway provision is Section 63(4) BSA , which requires a certificate signed by a person occupying a responsible position in relation to the operation of the relevant device or the management of the relevant activities. This certificate must identify the electronic record, describe how it was produced, give particulars of the device involved, and confirm that the device was operating properly. Without this certificate, the video is generally inadmissible as secondary electronic evidence — a principle the Supreme Court established conclusively in Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal (2020) 7 SCC 1. Landmark Ruling Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal , (2020) 7 SCC 1 — The Supreme Court held that the certificate under Section 65B(4) of the Indian Evidence Act (now Section 63(4) BSA) is a mandatory condition precedent to the admissibility of secondary electronic evidence. Without it, the evidence cannot be admitted regardless of its apparent probative value. Read on Indian Kanoon → However — and this is the crucial gap — the Section 63 certificate attests to the production and integrity of the copy , not to the authenticity of the underlying content . A certificate can confirm that a video file was accurately copied from a particular computer, but it cannot confirm that what the video depicts actually happened. A deepfake video can pass the Section 63 test if it was stored on a computer that was operating properly and the copy was accurately made. The fabrication lies in the content, not in the electronic process of storage and retrieval. This is precisely why the admissibility framework alone is insufficient for the deepfake era. Admissibility gets a video through the courtroom door; authenticity determines whether the court should believe what it shows. These are separate legal inquiries, and Indian courts are only beginning to develop frameworks for the latter in the context of AI-generated media. 6. How Indian Courts Examine Digital Evidence How do courts verify video evidence in India? The process involves multiple layers of scrutiny — statutory, procedural, and forensic — each designed to filter out unreliable material before a judge relies on it for a verdict. 6.1 The Section 63 BSA Certificate The first and most fundamental check is the mandatory certificate under Section 63(4) of the Bharatiya Sakshya Adhiniyam. As established in Anvar P.V. v. P.K. Basheer (2014) 10 SCC 473 and reaffirmed in Arjun Panditrao Khotkar (2020), this certificate is not a formality — it is a condition precedent without which secondary electronic evidence cannot cross the admissibility threshold. The certificate must identify the electronic record, describe the manner of its production, give particulars of the device, and confirm proper operation during the relevant period. 6.2 Chain of Custody Courts examine the chain of custody — the documented trail of how the video moved from the point of recording to the courtroom. Every transfer, from the original capture device through any messaging platforms, cloud storage, forensic laboratories, and finally the court filing, must be accounted for. Any undocumented gap creates an opportunity for tampering and gives the opposing party grounds to challenge the evidence. In deepfake cases, the chain-of-custody argument becomes especially powerful: the defence can argue that the original recording was replaced with an AI-generated fabrication at any undocumented point in the chain. 6.3 Expert Testimony and Forensic Analysis Section 39 of the BSA, 2023 permits courts to rely on the opinions of experts when the court needs to form an opinion on a point of science, art, or technology. For the purposes of electronic evidence, Section 79A of the IT Act, 2000 empowers the Central Government to appoint an Examiner of Electronic Evidence — a qualified expert whose opinion carries special weight. India's Central Forensic Science Laboratory (CFSL) and state-level Directorate of Forensic Science Laboratories (DFSLs) conduct forensic examinations of digital evidence, including video authentication. Forensic examination of video evidence typically includes metadata analysis (examining timestamps, device information, GPS coordinates, and file headers), hash verification (comparing SHA-256 cryptographic hash values to confirm the file has not been altered), compression artefact analysis (examining encoding patterns for signs of re-processing or splicing), and AI-based detection (using specialised tools that identify generative adversarial network signatures, facial inconsistencies, unnatural blinking patterns, lighting mismatches, and audio-visual synchronisation anomalies characteristic of deepfakes). 6.4 The IT Rules 2026 Provenance Layer With the IT Rules 2026 now in effect, a new evidentiary layer is emerging. Under Rule 3(3)(a)(ii), AI tool providers must embed permanent metadata identifiers in synthetic content at the point of generation, and visual labels must cover at least 10% of the screen area. Under Rule 4(1A), Significant Social Media Intermediaries (SSMIs) must obtain pre-publication declarations from uploaders and maintain logs of SGI content. These platform-side provenance records — including upload timestamps, account metadata, and SGI markers — are now available as additional evidence in court proceedings. As the Mondaq analysis of March 2026 observed, "failure to request or preserve such provenance data early in an investigation can irreversibly weaken the chain of custody." 7. What Laws Apply — The Complete Legal Framework What laws apply to deepfakes in India? As of May 2026, India does not have a single standalone statute dedicated exclusively to deepfakes. Instead, the legal framework is built from multiple overlapping statutes — the IT Act 2000, the Bharatiya Nyaya Sanhita 2023, the Bharatiya Sakshya Adhiniyam 2023, the Bharatiya Nagarik Suraksha Sanhita 2023, the Digital Personal Data Protection Act 2023, and the IT Rules 2026. Together, they create a layered regime that addresses deepfakes from the angles of evidence, criminal liability, platform accountability, and data protection. Statute / Provision What It Covers Maximum Punishment Section 66C, IT Act 2000 Identity theft — fraudulent use of electronic signature, password, or any unique identification feature 3 years + ₹1 lakh fine Section 66D, IT Act 2000 Cheating by personation using computer resource or communication device 3 years + ₹1 lakh fine Section 66E, IT Act 2000 Capturing, publishing, or transmitting images of a person's private area without consent 3 years + ₹2 lakh fine Section 67, IT Act 2000 Publishing or transmitting obscene material in electronic form First conviction: 3 years + ₹5 lakh fine; Subsequent: 5 years + ₹10 lakh fine Section 67A, IT Act 2000 Publishing or transmitting sexually explicit material in electronic form First conviction: 5 years + ₹10 lakh fine; Subsequent: 7 years + ₹10 lakh fine Section 228, BNS 2023 Fabricating false evidence — creating false documents, electronic records, or circumstances to mislead proceedings Per Section 229 (in judicial proceedings): 7 years + ₹10,000 fine Section 229, BNS 2023 Punishment for intentionally giving or fabricating false evidence In judicial proceedings: 7 years + fine; Other cases: 3 years + fine Section 336, BNS 2023 Forgery — making a false document or electronic record with intent to cause harm or support a claim 2 years + fine; for court records/public registers: up to 7 years Section 340, BNS 2023 Forgery for purpose of cheating 7 years + fine Section 351, BNS 2023 Criminal intimidation 2 years + fine (7 years if threat of death/grievous hurt) Section 353, BNS 2023 Statements conducing to public mischief — spreading false information causing fear, alarm, or communal enmity 3 years + fine Section 356, BNS 2023 Defamation 2 years + fine Section 77, BNS 2023 Voyeurism — capturing/disseminating images of a woman in a private act First offence: 1–3 years + fine; Subsequent: 3–7 years + fine Section 63, BSA 2023 Admissibility of electronic records — mandatory certificate for secondary electronic evidence N/A (evidentiary threshold) Section 379, BNSS 2023 Procedure for courts to initiate prosecution for false evidence (replaces Section 340 CrPC) N/A (procedural mechanism) IT Rules 2026 (effective Feb 20, 2026) 3-hour takedown for court/govt-flagged deepfakes; 2-hour for NCII; 10% labelling; metadata provenance; SSMI obligations Loss of Section 79 safe harbour; platform liability DPDP Act, 2023 Processing biometric/personal data without consent — covers facial recognition extraction and voice biometrics used in deepfakes Up to ₹250 crore for significant violations Constitutional Foundation K.S. Puttaswamy v. Union of India , (2017) 10 SCC 1 — The Supreme Court recognised informational privacy as a fundamental right under Article 21. The court observed that "dangers to privacy in an age of information can originate not only from the State but from non-state actors as well," creating the constitutional basis for regulating deepfakes that violate personal data and likeness. Read on Indian Kanoon → 8. What If Someone Submits Fake Evidence? What is the punishment for submitting fake evidence in court in India? The consequences are severe — both criminally and procedurally. The Bharatiya Nyaya Sanhita, 2023, dedicates an entire chapter (Chapter XIV) to "False Evidence and Offences Against Public Justice," and the provisions apply squarely to anyone who submits a deepfake video as genuine evidence. Section 227 of the BNS defines "giving false evidence" — a person who, being legally bound by an oath or by law to state the truth, makes a false statement that they know or believe to be false, or that they do not believe to be true, is said to give false evidence. Section 228 defines "fabricating false evidence" — whoever causes any circumstance to exist, or makes any false entry in any book, record, or electronic record, or makes any document or electronic record containing a false statement, intending that such circumstance, entry, or statement may appear in evidence in a judicial or other proceeding and cause an erroneous opinion to be formed, is said to fabricate false evidence. The punishment framework under Section 229 BNS operates on a two-tier basis. If the false evidence is given or fabricated in a judicial proceeding, the accused faces imprisonment which may extend to seven years and a fine of ₹10,000. If the false evidence is given in any other proceeding, the imprisonment extends to three years with fine. Crucially, if the false evidence leads to a conviction for an offence punishable with imprisonment, the person who gave the false evidence faces a punishment proportional to the harm caused — potentially extending to the same punishment the wrongly convicted person received. The procedural mechanism for initiating prosecution is Section 379 of the Bharatiya Nagarik Suraksha Sanhita (BNSS), 2023 — which replaces the former Section 340 of the CrPC. Under this provision, any court that finds that a person has given false evidence or fabricated false evidence may, after recording its finding, make a written complaint and refer the matter to a Magistrate of the first class for prosecution. The Supreme Court has held that the power under this section must be exercised judiciously but firmly where the integrity of judicial proceedings is at stake. For someone who submits a deepfake video as genuine evidence, this means the act of submission itself — regardless of whether the court ultimately relies on it — constitutes the offence of fabricating false evidence under Sections 228–229 BNS, the separate offence of forgery under Section 336 BNS (since a deepfake is a false electronic record made with intent to cause damage or support a claim), and potential contempt of court if the fabrication is designed to obstruct or undermine the administration of justice. 9. What If Someone Creates or Distributes a Deepfake? Is creating a deepfake illegal in India? Yes — even if the deepfake is never submitted as court evidence, its creation and distribution are punishable under multiple provisions of existing law. The legal exposure depends on the type of deepfake and the purpose for which it was created. If the deepfake involves identity theft — using someone else's electronic signature, password, or unique identification feature — Section 66C of the IT Act applies, carrying imprisonment up to 3 years and fine up to ₹1 lakh. If the deepfake is used to cheat or deceive any person by impersonation — say, a video of a bank official instructing a victim to transfer funds — Section 66D applies with identical punishment. If the deepfake captures, publishes, or transmits images of a person's private area without consent, Section 66E applies (3 years plus minimum ₹2 lakh fine). For sexually explicit deepfakes, Section 67A applies with imprisonment up to 5 years and fine up to ₹10 lakh on first conviction. Under the BNS 2023 , the creation of a deepfake can constitute defamation under Section 356 (if it harms reputation), criminal intimidation under Section 351 (if it threatens harm), voyeurism under Section 77 (if it involves non-consensual intimate imagery of a woman), and public mischief under Section 353 (if it spreads false information causing fear, alarm, or communal enmity). For organised or coordinated deepfake campaigns, Section 111 of the BNS recognises organised cyber crime. The IT Rules 2026 — Platform-Level Accountability Since February 20, 2026, the IT (Intermediary Guidelines) Amendment Rules, 2026 have established India's first comprehensive regulatory framework for synthetic content. The key provisions include the formal definition of "synthetically generated information" (SGI) under Rule 2(1)(wa); a mandatory 3-hour takedown window for deepfakes flagged by court or government order (reduced from 36 hours); a 2-hour window specifically for non-consensual intimate deepfake imagery; mandatory AI labelling covering at least 10% of visual screen area or the first 10% of audio duration; permanent metadata embedding by AI tool providers that cannot be removed or suppressed by users; pre-publication user declarations on SSMI platforms about whether content is synthetically generated; and loss of Section 79 safe harbour for platforms that fail to comply. The Digital Personal Data Protection Act, 2023 adds another dimension. Deepfakes that process biometric data — facial features, voice prints, iris patterns — without meaningful consent violate the DPDP Act's core principles. The penalty for significant violations can reach ₹250 crore, making this one of the most financially consequential exposure points for organisations involved in deepfake creation or distribution. Latest Regulatory Action IT (Intermediary Guidelines) Amendment Rules, 2026 — Notified on February 10, 2026; effective from February 20, 2026. India became the first major economy to mandate quantitative labelling thresholds (10% of visual area) for AI-generated content and to impose sub-4-hour takedown requirements for flagged deepfakes. Read detailed analysis → 10. Landmark Cases & Real-World Examples What are the famous deepfake cases in India? The Indian judiciary has confronted deepfake technology across multiple dimensions — from matrimonial disputes to celebrity impersonation to electoral manipulation. The following cases represent the most significant judicial encounters with deepfake and AI-manipulated evidence as of May 2026. 10.1 Nirmaan Malhotra v. Tushita Kaul (Delhi HC, 2024) In this matrimonial maintenance dispute, the husband challenged the alimony awarded by the Family Court, contending that his wife was living in adultery and producing photographs as proof. The Delhi High Court refused to accept the photographs at face value, observing that courts now function in an "era of deepfakes" and that the burden of proving the authenticity of photographic evidence squarely rests on the party presenting it. The court held that the photographs could not be relied upon without independent forensic verification. This case represents the first explicit judicial acknowledgement in India that deepfake technology has fundamentally altered how courts must assess photographic and video evidence in everyday litigation — not just high-profile cases. 10.2 Sadhguru Jagadish Vasudev v. Igor Isakov (Delhi HC, 2025) In this landmark personality-rights case, the defendants operated rogue websites using AI-morphed deepfakes of a prominent yoga guru's voice, image, and distinctive appearance to run scam cryptocurrency and health-product endorsements. Justice Saurabh Banerjee issued an unprecedented "dynamic+" injunction — an order that protects the plaintiff's entire persona (name, image, likeness, voice) from AI misuse, with a 36-hour takedown schedule for platforms and a direction to YouTube and similar services to deploy automated detection technology to identify and remove identical infringing content. The ruling was pathbreaking because it recognised that static, URL-specific injunctions are inadequate for pandemic-scale digital harms that replicate across platforms. 10.3 Rashmika Mandanna Deepfake Case (Delhi Police, 2023–2024) In November 2023, a 14-second deepfake video showing actress Rashmika Mandanna entering an elevator went viral. The original video featured a British-Indian woman named Zara Patel; the deepfake had digitally replaced her face with Mandanna's. Delhi Police traced and arrested a 24-year-old B.Tech graduate from Andhra Pradesh's Guntur district under IT Act Sections 66C and 66E. The accused told police he had created the deepfake to gain followers on Instagram. This case was pivotal in demonstrating that deepfake prosecution was practically achievable under existing Indian law — and it became one of the public triggers for the regulatory push that culminated in the IT Rules 2026. 10.4 Aman Gupta Personality Rights Order (Delhi HC, May 2026) In May 2026, the Delhi High Court granted interim protection to boAt co-founder Aman Gupta after sexually explicit AI-generated deepfakes of his likeness were circulated online. The court ordered the immediate takedown of the content, recognising that deepfakes using a person's likeness for obscene purposes violate fundamental personality rights under Article 21 of the Constitution. The order extended protection to Gupta's face, voice, name, and catchphrases from any unauthorised AI-generated use. 10.5 Shashi Tharoor Deepfake Order (Delhi HC, May 2026) In parallel proceedings in May 2026, Congress MP Shashi Tharoor moved the Delhi High Court against AI-generated deepfake videos that showed him making statements he never made. The court issued notice and indicated interim relief, ordering the blocking and removal of the fabricated content and recognising the grave threat deepfakes pose to the democratic process and individual reputation. The case underscored that deepfake threats extend far beyond entertainment or fraud — they strike at the heart of political discourse and electoral integrity. 10.6 Gujarat High Court — PIL on AI Misuse (April 2026) In April 2026, the Gujarat High Court issued notices to Meta, Google, X, Reddit, and Scribd on a public interest litigation seeking stricter enforcement of regulations against AI-generated deepfakes. The court called for strict compliance with the newly notified IT Rules 2026, signalling that High Courts across India — not just Delhi — are treating deepfake regulation as a matter of urgent judicial concern. The PIL sought a comprehensive regulatory framework addressing the gap between the IT Rules' requirements and ground-level enforcement. 10.7 Supreme Court Disposes PIL on Deepfake Framework (2026) The Supreme Court of India disposed of a PIL seeking a regulatory framework specifically for deepfakes after the Centre informed the court that draft AI rules had already been framed and notified — referring to the IT Rules 2026. While the disposal signals that the court considers the legislative response adequate for now, practitioners expect the constitutional validity of the 3-hour takedown rule and the prevention obligations to be tested in fresh proceedings within the next 12–18 months. 10.8 Supreme Court Flags AI-Generated Fake Judgments (2025–2026) In a separate but related development, the Supreme Court strongly condemned the use of AI-generated fake case laws and judgments in actual court proceedings. A trial court in Andhra Pradesh was found to have adjudicated a property dispute citing AI-generated orders that did not exist. The Supreme Court warned of legal consequences for such practices, directed the Bar Council to take action, and issued guidelines for courts to verify the authenticity of cited judgments. While this did not involve deepfake video, it demonstrated the judiciary's growing awareness that AI fabrication threatens the evidentiary foundation of legal proceedings. Stay Updated on Landmark Court Rulings Our curated Judgement Roundups bring you the most significant Indian court decisions — distilled, analysed, and explained by legal experts. Never miss a ruling that shapes the law. Read Our Judgement Roundups → 11. How Police, Lawyers & Courts Verify Videos How do courts verify video evidence? Can edited CCTV footage be trusted? The verification of digital video evidence in India follows a structured process that combines statutory requirements, forensic science, and increasingly, AI-based detection tools. The process operates at three levels — the investigative stage, the pre-trial stage, and the trial stage. 11.1 At the Investigation Stage When police seize a device containing video evidence, the first step is to generate a cryptographic hash value (typically SHA-256) of the original file immediately upon seizure. This hash serves as a digital fingerprint — any subsequent alteration to the file, even a single bit, will produce a different hash. The device and the file are then sent to a forensic laboratory — either the Central Forensic Science Laboratory (CFSL), a state-level Directorate of Forensic Science Laboratory (DFSL), or, under Section 79A of the IT Act, a government-notified Examiner of Electronic Evidence. The forensic examiner analyses the file's metadata (including creation date, modification history, device identifiers, and GPS data), examines compression artefacts for signs of re-encoding or splicing, and — increasingly — runs the file through AI-based deepfake detection tools that look for generative adversarial network (GAN) signatures, facial geometry anomalies, unnatural micro-expressions, and audio-visual synchronisation defects. 11.2 At the Pre-Trial Stage Before the video can be tendered as evidence, the party relying on it must prepare a Section 63(4) BSA certificate. The lawyer or investigating officer must also be prepared to establish the chain of custody — documenting every transfer of the file from seizure to court filing. Under the IT Rules 2026, if the video passed through any social media platform, the party can now request platform-side provenance data, including upload timestamps, account metadata, and SGI labels or markers that indicate whether the platform's automated systems flagged the content as synthetic. 11.3 At Trial At trial, the opposing party has the right to cross-examine the person who issued the Section 63 certificate and to challenge the forensic expert's report. The court may also, under Section 39 of the BSA, appoint its own independent expert if it considers the existing forensic evidence insufficient or contested. In cases where deepfake fabrication is alleged, the court can order a fresh forensic examination by a government-notified Examiner of Electronic Evidence, whose opinion carries special statutory weight. Evidence Verification Checklist — For Lawyers & Litigants ✓ Section 63(4) BSA Certificate — Obtain and verify the mandatory certificate from the person responsible for the relevant device or system. Without it, the video is generally inadmissible. ✓ Cryptographic Hash Verification — Record SHA-256 hash values at the earliest possible stage and compare against the file presented in court. Any discrepancy indicates tampering. ✓ Chain of Custody Documentation — Map every transfer from original device → forensic lab → court filing. Document dates, times, personnel, and methods at each step. ✓ Metadata Analysis — Examine timestamps, device identifiers, GPS data, file headers, and modification history for inconsistencies. ✓ Forensic Expert Report — Engage a government-notified Examiner of Electronic Evidence (under Section 79A, IT Act) or CFSL/DFSL for comprehensive analysis. ✓ AI-Based Deepfake Detection — Request forensic lab to run detection tools analysing GAN signatures, facial geometry, micro-expression anomalies, and audio-visual sync. ✓ Platform Provenance Data (IT Rules 2026) — Request SGI labels, upload logs, and automated detection flags from the intermediary platform under the new regulatory framework. ✓ Cross-Examination Strategy — Prepare to challenge the certificate, the chain of custody, the forensic methodology, and the qualifications of any expert witness. 12. What Citizens Should Do What should a citizen do if they discover a deepfake of themselves? What steps should you take if you suspect fake video evidence is being used against you in court? The answer in both situations is: act immediately and preserve everything. 12.1 If You Discover a Deepfake of Yourself Online The first and most critical step is evidence preservation. Before the content is taken down — and before the creator can delete it — take full-page screenshots with visible timestamps and URLs, save archived copies using web archiving services such as the Wayback Machine or Archive.today, download the deepfake file itself if possible and record its SHA-256 hash, and photograph or screen-record the platform's engagement metrics (likes, shares, comments, view counts). Next, file a complaint on the National Cyber Crime Reporting Portal at cybercrime.gov.in or call the helpline at 1930 . Simultaneously, send IT Rules 2026 takedown notices to the platform's designated grievance officer. Under the amended rules, platforms are required to remove non-consensual intimate imagery and impersonation deepfakes within 2 hours, and other flagged deepfakes within 3 hours of a lawful order. File an FIR at your local police station citing the relevant IT Act sections (66C, 66D, 66E) and BNS provisions. For urgent interim relief, consult a lawyer about filing an ex-parte injunction in the appropriate High Court. The Delhi and Bombay High Courts have demonstrated a clear willingness to grant interim injunctions against deepfakes within 24 to 72 hours, including "dynamic+" injunctions that cover re-uploads and mirror sites. In May 2026, the Delhi High Court granted such orders in favour of entrepreneur Aman Gupta and Congress MP Shashi Tharoor within days of filing. 12.2 If Fake Video Evidence Is Used Against You in Court If you believe a fake or manipulated video has been submitted as evidence against you in a court proceeding, raise the authenticity challenge at the earliest possible stage. Your lawyer should file an application challenging the admissibility and authenticity of the video, demanding production of the Section 63(4) BSA certificate, requesting full chain-of-custody documentation, seeking an independent forensic examination by a government-notified Examiner of Electronic Evidence, and requesting platform-side provenance data under the IT Rules 2026 framework. Your lawyer should also file an application under Section 379 BNSS requesting the court to initiate proceedings against the person who submitted the fake evidence for the offence of fabricating false evidence under Sections 228–229 BNS. Need Legal Help with Digital Evidence or Cyber Crime? Whether you're a victim of deepfake abuse, facing fabricated evidence, or need expert consultation on electronic evidence law — The EduLaw's team of legal professionals is here to help. Book a Legal Consultation → 13. The Future of Evidence in India Where is the law heading on deepfake evidence in India? The trajectory is clear — Indian law is moving toward a future where digital evidence faces significantly higher scrutiny, where AI tools play a dual role as both the threat and the detection mechanism, and where the evidentiary framework will need to be substantially updated to keep pace with synthetic media technology. The IT Rules 2026 represent a critical first step, but they are not the final word. Several developments are expected in the near term. First, the constitutional validity of the 3-hour takedown rule and the prevention obligations will almost certainly be tested before the Supreme Court within the next 12–18 months, with petitioners arguing that the compressed timelines create over-removal incentives that chill legitimate speech under Article 19(1)(a). Second, MeitY is expected to publish approved technical standards for deepfake detection and provenance metadata, giving courts a clearer benchmark against which to measure the reliability of forensic evidence. Third, C-DAC's AI detection toolchain, currently in field-test stages, is expected to move to broader deployment, giving forensic laboratories across India access to standardised detection capabilities. The proposed Digital India Act , which is expected to eventually replace the IT Act, 2000, will likely include dedicated provisions for AI-generated evidence and synthetic media — moving beyond the current framework of layered provisions across multiple statutes toward a more unified regime. Legislative proposals such as the Deepfake Prevention and Criminalisation Bill (introduced in the Rajya Sabha in 2023, though not yet enacted) signal Parliamentary appetite for standalone deepfake legislation. On the judicial front, courts are beginning to develop deepfake-specific evidentiary standards. The practice of ordering independent forensic examination in contested cases, the recognition that visual evidence alone is insufficient without corroboration, and the judicial notice that courts operate in "an era of deepfakes" are all laying the doctrinal foundation for what may eventually become formal evidentiary guidelines for AI-generated media. For citizens, the fundamental message is this: the law is catching up, but it has not yet caught up. In the interim, the burden falls on individuals to preserve evidence, assert their rights quickly, and ensure that their lawyers understand both the legal framework and the technology. The gap between deepfake creation capability and deepfake detection capability is the space where injustice thrives — and closing that gap is the single most urgent task facing India's legal system. 14. Key Takeaways Summary — What You Must Know Yes, a fake video can send someone to jail — if it is submitted as evidence and not detected as fabricated before the court relies on it. The risk is real and growing as deepfake technology becomes more accessible. Indian law treats vi