TaxTMI 
TaxTMI 
TaxTMI 
Here’s a detailed and structured analysis of Artificial Intelligence (AI) in the context of Indian Cyber Lawsvis-à-visIndian Intellectual Property Rights (IPR) Laws, with an emphasis on overlaps, conflicts, and regulatory gaps:
I. OVERVIEW: Why this Intersection Matters
AI is increasingly:
- Creating content (text, music, art, code),
- Making autonomous decisions (e.g., facial recognition, predictive policing),
- Being deployed across sectors (finance, education, health),
- Raising questions about ownership, liability, infringement, and privacy.
Key Indian Legal Domains Involved:
Domain | Key Laws / Authorities |
Cyber Laws | IT Act, 2000 (incl. amendments + IT Rules 2021/2023), CERT-In Guidelines, DPDP Act, 2023 |
IPR Laws | Copyright Act, 1957; Patents Act, 1970; Trademarks Act, 1999; Designs Act, 2000; GI Act |
Overlapping Areas | AI-generated works, data scraping, deepfakes, voice cloning, content moderation, etc. |
II. INDIAN CYBER LAWS & AI
1. Information Technology Act, 2000 (IT Act)
Though the IT Act predates modern AI, it governs the digital ecosystem in which AI operates.
Relevance to AI:
Issue | How IT Act Applies |
AI-generated deepfakes / impersonation | Sec 66D (cheating by impersonation), Sec 67 (obscenity), Sec 66E (violation of privacy) |
Autonomous decision-making AI | No direct regulation; liability may fall on deploying entity |
Algorithmic bias / discrimination | No legal recognition yet — needs future AI-specific rules |
AI bots & misinformation | Covered under IT Rules, 2021/2023 (esp. obligations on intermediaries) |
Cybersecurity in AI systems | CERT-In Guidelines (2022) mandate breach disclosure, logs, etc. |
Data Protection:
- Digital Personal Data Protection Act (DPDP), 2023:
AI that uses personal data must comply with consent, purpose limitation, and data minimization.- Personal data used for model training may violate DPDP if consent isn't obtained.
- DPDP doesn't yet address non-personal data, which is vital for training AI.
2. IT Rules (Intermediary Guidelines), 2021 / 2023
- Platforms must act against deepfake content, impersonation, misinformation (under Rule 3(1)(b)).
- AI-generated content that violates public order, decency, or national security can be removed.
- The upcoming Digital India Act (to replace IT Act) is expected to address AI ethics, algorithmic accountability, and explainability.
III. INDIAN INTELLECTUAL PROPERTY LAWS & AI
The big legal question: Can AI be an author/inventor? If not, who owns the IP?
1. Copyright Act, 1957
Issue | Legal Position (India) |
AI-generated art, music, text, code | No explicit provision for non-human authorship |
Who owns AI-generated work? | Currently, only a natural/legal person (human or company) can be author |
Derivative works using training data | May amount to infringement, esp. if original data was protected and not licensed |
Voice cloning / deepfake mimicry | Not copyright infringement per se, but may trigger personality rights or passing off |
Note: India follows a human-centric model. Unlike UK (which allows “computer-generated works” with a notional human author), India has not recognized AI as an autonomous creator.
2. Patent Act, 1970
Issue | Legal Position (India) |
Can AI be an inventor? | NO. Patent Rules require natural person to be inventor |
AI-assisted inventions | Patentable if a human inventor can be identified and explains the inventive step |
Algorithmic inventions / ML models | Software per se is not patentable, unless tied to hardware (as per Sec. 3(k)) |
AI as inventor has been debated worldwide (see DABUS case). Indian Patent Office has rejected AI inventorship claims.
3. Trademarks Act, 1999
- AI-generated logos, brand names can be registered, but only if filed by a human or legal entity.
- AI models used to auto-generate product names may create confusion or unintentional similarity ? potential for infringement / passing off.
- Use of AI in automated brand monitoring is permitted.
4. Designs Act, 2000
- Similar to copyright: human authorship required.
- AI-generated product designs or UIs may not qualify for registration if a human author cannot be attributed.
IV. CYBER LAW vs. IPR LAW — Comparison Table
Aspect | Cyber Law (IT Act, etc.) | IPR Law (Copyright, Patent, etc.) |
Objective | Regulate online conduct, data, cybercrimes | Protect creative and inventive output |
AI focus | Regulates use/misuse of AI tools | Regulates output from AI tools |
Ownership | Focuses on liability, consent, data control | Focuses on ownership, originality, authorship |
Harm addressed | Privacy violation, fraud, hacking, impersonation | Misuse of original work, plagiarism, piracy |
AI-generated content | May be restricted under public order/misuse grounds | Ownership unclear if no human author |
Voice cloning / Deepfakes | IT Act can penalize misuse / impersonation | Copyright doesn’t apply to voice/image per se |
V. Legal Gaps & Challenges
1. AI as Creator / Inventor: No Legal Recognition
- Indian laws do not recognize AI as a legal person ? no rights or liabilities can be directly assigned.
- No mechanism for assigning authorship of AI-generated content without human intervention.
2. Data Use for AI Training
- No law explicitly governs the use of copyrighted or personal data for training AI.
- DPDP Act only covers personal data, not public or anonymized data.
- No “text and data mining” exception as in EU copyright law.
3. Deepfakes & Synthetic Media
- India lacks a specific deepfake or voice cloning law.
- Personality/publicity rights are court-developed, not statutory.
4. AI Bias, Algorithmic Discrimination
- No mandatory auditing or transparency for AI algorithms under either IT or IPR laws.
- No guidelines on explainability, ethics, or harmful algorithmic outcomes.
VI. Future Outlook & Reforms
Anticipated Developments:
Area | Expected Changes |
Digital India Act (proposed) | AI governance, algorithmic transparency, platform accountability |
AI Policy / Guidelines (NITI Aayog / MeitY) | Draft principles for ethical AI (fairness, accountability, safety) |
New IPR Frameworks | Possible tweaks to handle AI-generated works and machine-invented patents |
Criminalization of Deepfakes | Proposed under future IT Act overhaul or criminal law reforms |
Global Influence:
- India may borrow models from:
- UK (AI-generated works allowed, with minimal human intervention)
- EU AI Act (risk-based classification, prohibited AI uses)
- WIPO (discussions on AI and IP underway)
- US (cases like Stephen Thaler v. USPTO on AI inventorship)
VII. Recommendations & Way Forward
To bridge the gap between AI, cyber law, and IP, India could consider:
- Amending Copyright & Patent Laws
- Define how AI-generated works can be protected.
- Assign ownership to developers, users, or platform by default where appropriate.
- Recognizing Digital Persona Rights
- Statutory recognition of personality/publicity rights, especially for voice and likeness.
- Deepfake-specific criminal law with consent-based thresholds.
- AI Governance Frameworks
- Require transparency, accountability, and explainability in AI systems.
- Ethical auditing of AI algorithms that impact public outcomes (e.g., in education, policing, healthcare).
- TDM (Text & Data Mining) Provisions
- Allow regulated use of data for model training with licensing / opt-out provisions.
- New Dispute Mechanisms
- Fast-track IP and cyber-related disputes involving AI (e.g., via IPAB-style tribunals or tech courts).
CONCLUSION
AI sits at the intersection of cyber laws (regulating digital conduct) and IP laws (regulating output and creativity). India currently handles AI issues in a **fragmented
, reactive** manner through general principles. However, as AI becomes central to both innovation and misuse (e.g. deepfakes, voice cloning), a cohesive legal framework is urgently needed.
The evolution of cases like Amitabh Bachchan’s personality rights, upcoming Digital India Act, and potential reforms to copyright/patent law are signals of progress — but legislative clarity, judicial consistency, and policy foresight are essential.