AI & Technology Disputes & Enforcement in India

AI and technology disputes in India are no longer peripheral legal events - they are central business risks for any company operating in the country';s digital economy. India';s legal framework for resolving AI-related conflicts spans multiple statutes, tribunals, and regulatory bodies, and the choice of enforcement pathway determines both the timeline and the commercial outcome. This article provides a structured guide to the legal tools available, the procedural realities on the ground, and the strategic decisions that international businesses must make before a dispute escalates.

The Indian technology sector is governed by a layered architecture of legislation: the Information Technology Act, 2000 (IT Act), the Copyright Act, 1957, the Patents Act, 1970, the Trade Marks Act, 1999, and the newly enacted Digital Personal Data Protection Act, 2023 (DPDPA). Each statute creates distinct rights, remedies, and enforcement mechanisms. Understanding which law governs a particular AI or technology dispute is the first decision a business must make, and an incorrect classification at the outset can result in wasted months and misdirected resources.

The article covers: the regulatory and judicial landscape for AI disputes; intellectual property enforcement for software, datasets, and AI models; contract and liability disputes in technology transactions; data protection enforcement under the DPDPA; and the role of arbitration as an alternative to court litigation.

India does not yet have a standalone AI-specific statute. Disputes involving artificial intelligence systems are resolved under the general framework of existing laws, supplemented by sector-specific regulations issued by bodies such as the Securities and Exchange Board of India (SEBI), the Reserve Bank of India (RBI), the Telecom Regulatory Authority of India (TRAI), and the Ministry of Electronics and Information Technology (MeitY).

The primary civil courts handling technology disputes are the High Courts of Delhi, Bombay, Madras, and Calcutta, each exercising original jurisdiction over intellectual property matters. The Delhi High Court, in particular, has developed a sophisticated body of jurisprudence on software copyright, domain name disputes, and technology contracts. The Intellectual Property Division (IPD) of the Delhi High Court, established under the Tribunals Reforms Act, 2021, now consolidates appeals from the erstwhile Intellectual Property Appellate Board (IPAB) and handles first-instance IP matters of national importance.

For disputes involving data protection violations, the DPDPA establishes a Data Protection Board of India (DPBI) as the primary adjudicatory authority. The DPBI is empowered to investigate complaints, impose financial penalties, and issue directions to data fiduciaries. Appeals from DPBI orders lie to the Telecom Disputes Settlement and Appellate Tribunal (TDSAT) and thereafter to the High Courts.

The IT Act, under Section 46, empowers an Adjudicating Officer appointed by the central government to adjudicate claims for compensation arising from contraventions of the Act, subject to a pecuniary limit. Matters exceeding that limit, or involving criminal liability, are referred to the Cyber Appellate Tribunal and thereafter to High Courts. Section 43A of the IT Act imposes liability on body corporates that fail to implement reasonable security practices, making it directly relevant to AI systems that process sensitive personal data.

A non-obvious risk for international companies is the jurisdictional complexity created by this multi-body structure. A single AI-related incident - for example, an algorithmic decision that causes financial harm and involves personal data - can simultaneously trigger proceedings before the DPBI, the Adjudicating Officer under the IT Act, and a civil court. Coordinating responses across these forums requires deliberate legal strategy from the outset.

Intellectual property is the most frequently contested category in Indian AI and technology disputes. The three primary IP instruments are copyright, patents, and trade secrets, each with distinct conditions of applicability.

Copyright protection for software and AI outputs. The Copyright Act, 1957, under Section 2(ffc), expressly includes computer programmes within the definition of literary works. This protection extends to source code, object code, and the structure, sequence, and organisation of a programme. However, Indian copyright law does not currently recognise AI-generated works as having an independent author - authorship must vest in a human creator. This creates a practical gap: outputs generated autonomously by an AI system may not attract copyright protection unless a human author can be identified as having made creative choices in the process.

For software copyright enforcement, the owner may seek an injunction, damages, or an account of profits under Section 55 of the Copyright Act. Interim injunctions are available under Order XXXIX of the Code of Civil Procedure, 1908 (CPC), and Indian courts have consistently granted ex parte injunctions in cases of clear software piracy, sometimes within 24 to 48 hours of filing. The standard for an interim injunction requires the applicant to demonstrate a prima facie case, balance of convenience, and irreparable harm - a threshold that is well-established in Delhi High Court practice.

Patent protection for AI-related inventions. The Patents Act, 1970, under Section 3(k), excludes mathematical methods, business methods, computer programmes per se, and algorithms from patentability. This exclusion has been interpreted broadly by the Indian Patent Office, making it difficult to obtain patents for AI algorithms in isolation. However, AI-enabled technical applications - such as a specific method of medical diagnosis using a neural network, or an AI-driven industrial process - may qualify for patent protection if the claimed invention produces a technical effect beyond the algorithm itself.

The Indian Patent Office has issued guidelines on computer-related inventions (CRI Guidelines) that provide a framework for assessing patentability. Applicants must demonstrate that the claimed invention, as a whole, produces a technical advancement and is not merely a software implementation of a known process. Prosecution timelines at the Patent Office can extend to three to five years for contested applications, and oppositions under Section 25 of the Patents Act add further delay.

Trade secrets and confidential information. India does not have a standalone trade secrets statute. Protection for confidential AI training data, proprietary algorithms, and model architectures relies on contractual confidentiality obligations and the equitable doctrine of breach of confidence, as applied by Indian courts. The absence of a statutory framework means that enforcement depends heavily on the quality of the underlying contracts and the speed with which injunctive relief is sought.

A common mistake made by international technology companies is failing to register copyright in India separately from their home jurisdiction. India is a member of the Berne Convention, which provides automatic protection, but local registration under Section 45 of the Copyright Act creates a public record and significantly strengthens the evidentiary position in litigation. Registration takes approximately 30 to 60 days for straightforward applications.

To receive a checklist for IP enforcement of AI assets in India, send a request to info@vlolawfirm.com.

Technology contracts in India - covering software development agreements, AI-as-a-service arrangements, data licensing agreements, and technology transfer agreements - are governed primarily by the Indian Contract Act, 1872. The enforceability of limitation of liability clauses, indemnity provisions, and service level agreements is determined by the general principles of contract law, supplemented by the Specific Relief Act, 1963.

Specific performance of technology contracts. The Specific Relief Act, 1963, as amended in 2018, significantly expanded the availability of specific performance. Under Section 10 of the amended Act, specific performance is now ordinarily available unless the contract is determinable in nature or involves personal service. For technology contracts - such as an obligation to deliver a functional AI system or to provide access to a proprietary dataset - specific performance is a viable remedy where monetary damages would be inadequate. The 2018 amendment removed the court';s discretion to refuse specific performance on grounds of hardship alone, making this remedy more predictable for technology businesses.

Liability for AI-caused harm. Indian law does not yet have a specific liability regime for harm caused by autonomous AI systems. Liability is assessed under the general law of torts, the Consumer Protection Act, 2019, and, where applicable, the IT Act. The Consumer Protection Act is particularly relevant where an AI product or service causes harm to an end consumer, as it provides for product liability under Chapter VI. A manufacturer, service provider, or seller of an AI-enabled product can be held liable for defects in design, manufacturing, or inadequate instructions.

The doctrine of strict liability, as established in Indian jurisprudence following the principle in M.C. Mehta v. Union of India, applies to enterprises engaged in hazardous or inherently dangerous activities. Whether AI systems deployed in high-stakes contexts - such as autonomous vehicles, medical diagnostics, or financial trading - would attract strict liability is an open question that Indian courts have not yet definitively resolved. This uncertainty is a material risk for businesses deploying AI in safety-critical applications.

Practical scenarios in contract disputes.

Consider three representative situations. First, a foreign software company contracts with an Indian enterprise to develop a machine learning model for credit scoring. The model underperforms against agreed benchmarks. The Indian client withholds payment and claims damages. The dispute turns on the interpretation of the performance specifications in the contract and whether the shortfall constitutes a fundamental breach. If the contract is silent on the standard of care, Indian courts apply the standard of reasonable skill and care under Section 69 of the Indian Contract Act.

Second, an Indian AI startup licenses its natural language processing technology to a multinational. The multinational uses the technology beyond the licensed scope, incorporating it into products not covered by the agreement. The startup seeks an injunction and damages for breach of contract and copyright infringement simultaneously. The overlap between contractual and IP remedies requires careful pleading to avoid inconsistent positions.

Third, a cloud services provider suffers a data breach affecting AI training datasets belonging to multiple clients. Each client suffers different categories of loss: some lose competitive advantage through exposure of proprietary data, others face regulatory penalties. The provider';s standard terms contain a limitation of liability clause capping liability at three months of fees. Indian courts have upheld such clauses where they are freely negotiated between commercial parties, but have struck them down where they are unconscionable or where the breach involves fraud or wilful misconduct under Section 23 of the Indian Contract Act.

A non-obvious risk in technology contracts governed by Indian law is the treatment of liquidated damages clauses. Under Section 74 of the Indian Contract Act, a party claiming liquidated damages must still prove actual loss - unlike many common law jurisdictions where a genuine pre-estimate of loss is enforceable without proof of actual damage. International businesses that rely on liquidated damages clauses as a substitute for proving loss will find this a significant limitation.

The Digital Personal Data Protection Act, 2023 (DPDPA) represents the most significant regulatory development affecting AI and technology businesses in India. The Act applies to the processing of digital personal data within India, and to processing outside India where it involves offering goods or services to individuals in India.

Key obligations for AI businesses. Under Section 4 of the DPDPA, personal data may be processed only for a lawful purpose for which the data principal has given consent, or for certain legitimate uses specified in the Act. AI systems that process personal data - including recommendation engines, facial recognition systems, and behavioural analytics platforms - must identify a lawful basis for each processing activity. Consent must be free, specific, informed, unconditional, and unambiguous, and must be obtained through a clear affirmative action.

Section 8 of the DPDPA imposes obligations on data fiduciaries to implement appropriate technical and organisational measures to ensure compliance. For AI systems, this translates into requirements for data minimisation, purpose limitation, and accuracy of data used in training and inference. The Act does not yet specify technical standards for AI systems, but MeitY is expected to issue sector-specific rules that will elaborate on these obligations.

Financial penalties under the DPDPA. The Act establishes a penalty framework under Schedule 1 that is among the most significant in Indian regulatory history. Penalties for failure to implement adequate security safeguards can reach INR 250 crore (approximately USD 30 million) per instance of breach. Penalties for failure to notify the DPBI of a personal data breach can reach INR 200 crore. These figures represent a material financial risk for any business operating AI systems that process personal data at scale.

Enforcement process. The DPBI investigates complaints filed by data principals or initiated suo motu. The Board has the power to summon documents, examine witnesses, and impose penalties after providing the respondent an opportunity to be heard. The procedural timeline for DPBI proceedings has not yet been established by regulation, but the Act contemplates a structured adjudication process with defined stages. Appeals from DPBI orders lie to TDSAT within 60 days of the order.

A common mistake for international businesses is assuming that data localisation requirements under the DPDPA are limited to sensitive personal data. The Act empowers the central government to notify categories of data that must be stored within India, and this notification power is broad. Businesses that have designed their AI infrastructure on the assumption of unrestricted cross-border data flows should review their architecture against the evolving notification framework.

To receive a checklist for DPDPA compliance for AI systems operating in India, send a request to info@vlolawfirm.com.

Arbitration is increasingly the preferred dispute resolution mechanism for technology disputes in India, particularly where the parties are from different jurisdictions or where confidentiality is a commercial priority. The Arbitration and Conciliation Act, 1996 (A&C Act), as amended in 2015, 2019, and 2021, provides the statutory framework.

Conditions for effective arbitration in AI disputes. An arbitration agreement must be in writing under Section 7 of the A&C Act. For technology contracts, this is typically satisfied by an arbitration clause in the master services agreement or the software licence. The clause should specify the seat of arbitration, the governing law, the number of arbitrators, and the institutional rules. Common choices for India-seated arbitrations include the Mumbai Centre for International Arbitration (MCIA), the Delhi International Arbitration Centre (DIAC), and the Indian Council of Arbitration (ICA). For disputes with a strong international dimension, parties often choose Singapore (SIAC) or London (LCIA) as the seat, with Indian law as the governing law.

Interim measures in technology arbitrations. Section 9 of the A&C Act allows a party to apply to a court for interim measures before or during arbitration. This is particularly valuable in technology disputes where urgent relief - such as an injunction against the use of misappropriated source code or the preservation of digital evidence - is needed before an arbitral tribunal is constituted. Courts have granted Section 9 relief in technology disputes within days of application where the urgency is demonstrated. Once the tribunal is constituted, Section 17 empowers the tribunal itself to grant interim measures, and court-ordered enforcement of such measures is available under Section 17(2).

Enforcement of foreign arbitral awards in India. India is a signatory to the New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards, 1958. Foreign awards are enforceable under Part II of the A&C Act. Enforcement proceedings are filed before the relevant High Court, and the grounds for resisting enforcement are limited to those specified in Section 48 of the Act, which mirror the New York Convention grounds. In practice, enforcement of foreign awards in India can take one to three years, depending on the complexity of the challenge and the court';s docket. The public policy ground for resisting enforcement has been narrowed by the 2015 amendment, which limits it to cases involving fraud, corruption, or a violation of the fundamental policy of Indian law.

Comparison of arbitration and court litigation for technology disputes. Court litigation in Indian High Courts offers the advantage of established precedent, the availability of ex parte injunctions, and the ability to join multiple parties. However, timelines for final judgment in complex technology disputes can extend to five to ten years in contested matters. Arbitration offers confidentiality, party autonomy in selecting arbitrators with technical expertise, and - for institutional arbitrations - more predictable timelines, typically 18 to 36 months for a final award. The cost of arbitration is generally higher in absolute terms than court filing fees, but the overall economic cost when accounting for management time and prolonged uncertainty often favours arbitration for disputes above a certain threshold.

Many international businesses underappreciate the importance of drafting the arbitration clause with precision. A poorly drafted clause - for example, one that fails to specify the seat or uses ambiguous language about the scope of disputes covered - can result in satellite litigation over the validity or scope of the clause before the substantive dispute is even addressed. Indian courts have spent considerable judicial time resolving disputes arising from defective arbitration clauses in technology contracts.

We can help build a strategy for structuring dispute resolution clauses in your India technology contracts. Contact info@vlolawfirm.com.

Effective enforcement in AI and technology disputes in India requires a structured approach that begins before any dispute arises. The following framework reflects the progression from legal context through tools to application, risks, and solutions.

Pre-dispute documentation and evidence preservation. Indian courts and arbitral tribunals place significant weight on contemporaneous documentary evidence. For AI disputes, this means maintaining detailed records of: the specifications agreed for an AI system; version histories of software and model weights; data provenance records for training datasets; communications between the parties during development and deployment; and audit logs of AI system outputs. The Indian Evidence Act, 1872, as amended by the Information Technology Act, governs the admissibility of electronic records. Section 65B of the Evidence Act requires a certificate from a responsible official to authenticate electronic evidence, and failure to comply with this requirement has resulted in the exclusion of critical evidence in Indian proceedings.

Pre-litigation notices and demand letters. Before initiating court proceedings or arbitration, it is standard practice in India to send a formal legal notice under Section 80 of the CPC (for suits against government bodies) or a demand letter in commercial disputes. For IP infringement, a cease-and-desist letter serves both as a formal demand and as evidence of the infringer';s knowledge, which is relevant to the assessment of damages. The notice period is typically 30 to 60 days, though urgent matters can proceed without notice where delay would cause irreparable harm.

Choosing the right forum. The choice between the High Court, the DPBI, the Adjudicating Officer under the IT Act, and arbitration depends on several factors: the nature of the right being enforced (IP, contractual, or regulatory); the urgency of the relief required; the value of the dispute; the need for confidentiality; and the enforceability of any order or award. For disputes involving both IP infringement and breach of contract, the High Court is typically the most efficient forum because it can grant relief on both grounds in a single proceeding. For disputes that are primarily contractual and where the parties have agreed to arbitration, the arbitral route is generally preferable.

Scenario: a foreign AI company facing infringement in India. A European company discovers that an Indian competitor has copied the architecture of its AI-powered analytics platform and is marketing it to Indian enterprises. The European company holds copyright in the software and has filed a patent application in India. The appropriate immediate step is to file for an interim injunction before the Delhi High Court, supported by evidence of the copyright ownership and the infringement. The court can grant an ex parte injunction restraining the competitor from marketing the product pending a full hearing. Simultaneously, the company should file a complaint with the Cyber Crime Cell of the relevant police jurisdiction, as software piracy is a cognisable offence under Section 63 of the Copyright Act.

Scenario: an Indian enterprise disputing an AI service contract. An Indian bank has contracted with a foreign AI vendor for a fraud detection system. The system generates a high rate of false positives, causing the bank to decline legitimate transactions and suffer reputational harm. The bank seeks to terminate the contract and recover damages. The vendor';s contract contains a limitation of liability clause and an arbitration clause specifying Singapore as the seat. The bank must assess whether the limitation clause is enforceable under Indian law (given the Section 74 requirement to prove actual loss), whether the arbitration clause covers the dispute, and whether to seek interim relief from an Indian court under Section 9 of the A&C Act while arbitration is pending.

Scenario: a startup facing a data protection investigation. An Indian AI startup operating a consumer-facing recommendation engine receives a notice from the DPBI following a complaint by a data principal. The DPBI requests information about the startup';s data processing activities, consent mechanisms, and security measures. The startup must respond within the time specified in the notice - typically 30 days - and provide detailed documentation of its compliance programme. Failure to respond or providing incomplete information can result in enhanced penalties. The startup should engage legal counsel immediately and conduct an internal audit of its data processing activities before responding.

The cost of inaction. Delay in initiating enforcement proceedings carries specific risks in India. For copyright infringement, the limitation period under the Limitation Act, 1963, is three years from the date of infringement. For contract disputes, the limitation period is also three years from the date of breach. Missing these deadlines extinguishes the right to sue, regardless of the merits of the claim. For interim injunctions, delay in filing after discovering the infringement weakens the argument of irreparable harm and can result in the court refusing interim relief on the ground that the applicant has acquiesced.

To receive a checklist for pre-dispute preparation and enforcement strategy for AI & technology disputes in India, send a request to info@vlolawfirm.com.

What is the most significant practical risk for a foreign company enforcing AI-related IP rights in India?

The most significant practical risk is the gap between obtaining an order and achieving effective enforcement. Indian courts can grant injunctions and damages awards, but execution of decrees against Indian defendants requires separate proceedings before the executing court, which can be slow. A defendant may continue infringing activity during the execution process, particularly if it has restructured its business to avoid the scope of the injunction. International companies should structure their enforcement strategy to include both injunctive relief and, where possible, criminal complaints under the Copyright Act, which can result in search and seizure of infringing materials and create stronger deterrence. Engaging local enforcement counsel with experience in coordinating civil and criminal proceedings simultaneously is essential.

How long does it typically take to resolve an AI or technology dispute in India, and what are the likely costs?

The timeline depends heavily on the forum and the nature of the dispute. An interim injunction from the Delhi High Court can be obtained within days to weeks in urgent cases, but a final judgment on the merits in a contested IP or contract dispute typically takes three to seven years in the High Court. Arbitration before an institutional body in India typically takes 18 to 36 months for a final award, and Singapore-seated arbitrations involving Indian parties tend to follow similar timelines. Legal fees for complex technology disputes in India start from the low tens of thousands of USD for straightforward matters and can reach several hundred thousand USD for multi-year High Court litigation or international arbitration. State court fees are calculated as a percentage of the claim value and can be substantial for high-value disputes. Businesses should factor these costs into their decision to pursue enforcement versus settlement.

When should a business choose arbitration over court litigation for an AI technology dispute in India?

Arbitration is preferable where confidentiality is important - for example, where the dispute involves proprietary AI model architectures or sensitive business data that would become part of the public court record in litigation. It is also preferable where the parties want to select arbitrators with technical expertise in AI or software, which is not possible in court litigation. Court litigation is preferable where urgent ex parte interim relief is needed immediately, where the dispute involves multiple parties who have not all agreed to arbitrate, or where the enforceability of the final order is a priority and the defendant';s assets are located in India (since court decrees are directly executable without a separate enforcement proceeding). For disputes above approximately USD 500,000 in value, the economics of institutional arbitration generally compare favourably to prolonged High Court litigation when the total cost of management time and uncertainty is included.

AI and technology disputes in India are governed by a complex, multi-layered legal framework that rewards advance preparation and penalises reactive enforcement. The combination of IP law, contract law, data protection regulation, and sector-specific oversight creates both risks and opportunities for businesses that understand the system. Choosing the right forum, preserving evidence correctly, and structuring contracts with enforceable dispute resolution clauses are the three decisions that most determine the outcome of a technology dispute in India.

Our law firm VLO Law Firms has experience supporting clients in India on AI and technology dispute matters. We can assist with IP enforcement strategy, contract dispute resolution, DPDPA compliance advisory, and arbitration clause drafting and representation. To receive a consultation, contact: info@vlolawfirm.com.