Executive Summary
As artificial intelligence increasingly generates contractual documents, Indian law faces novel questions about validity, enforceability, and liability allocation. While no specific Karnataka HC ruling exists on AI-drafted contracts as of January 2026, the legal framework can be analyzed through existing contract law principles and emerging judicial guidance on AI use in legal practice.
Key Findings:
- AI-generated contracts can be valid under Indian Contract Act if basic requirements are met
- Human review and approval creates binding intent
- Liability allocation depends on whether AI is a "tool" or "agent"
- Kerala HC's 2025 AI Policy sets precedent for AI use in legal contexts
Introduction
The integration of AI in contract drafting has moved from experimental to mainstream. Legal tech platforms now generate everything from employment agreements to complex commercial contracts. But when an AI drafts a contract that causes harm, who bears responsibility?
This article analyzes the enforceability of AI-generated contracts under Indian law and proposes a liability framework for when things go wrong.
Section 1: Legal Validity of AI-Generated Contracts
Requirements Under Indian Contract Act, 1872
For any contract to be valid under Section 10, it must have:
| Requirement | AI Contract Analysis |
|---|---|
| Offer and Acceptance | AI generates terms; humans accept/modify |
| Free Consent | Consent given by human parties, not AI |
| Competent Parties | Human parties must be competent |
| Lawful Consideration | Consideration determined by parties |
| Lawful Object | AI should not generate illegal terms |
The "Intent" Question
The fundamental question: Can a machine have contractual intent?
Analysis: Under Indian law, intent resides with the human parties who:
- Commission the AI to draft the contract
- Review the generated output
- Approve and execute the document
The AI is a sophisticated tool - like a word processor or template - rather than a contracting party.
IT Act Support
Section 10A of the Information Technology Act, 2000 validates contracts formed through electronic means:
"Where in a contract formation, the communication of proposals, the acceptance of proposals, the revocation of proposals and acceptances, as the case may be, are expressed in electronic form or by means of an electronic record, such contract shall not be deemed to be unenforceable solely on the ground that such electronic form or means was used for that purpose."
This supports the validity of AI-assisted contract formation.
Section 2: Emerging Judicial Guidance
Kerala High Court AI Policy (July 2025)
The Kerala High Court's Policy Regarding Use of AI Tools is India's first formal guidance on AI in legal contexts:
Key Provisions:
- Prohibition on using open-ended AI chatbots for court-related work
- Requirement for human verification of all AI-generated content
- Data privacy protections for case-related information
- Accountability framework for AI-assisted outputs
Relevance to Contracts: While focused on judicial work, the policy establishes principles applicable to legal drafting:
- Human oversight is mandatory
- AI outputs require verification
- Professionals remain accountable for AI-generated work
Delhi High Court's Warning (September 2025)
The Delhi High Court detected petitions using fake precedents generated by ChatGPT, warning of contempt proceedings. This reinforces that:
- AI hallucinations are a real risk in legal documents
- Professionals cannot blame AI for errors
- Verification obligations are non-negotiable
Section 2A: Judicial Precedents on Contract Validity and Enforceability
Indian courts have developed robust jurisprudence on contract validity that directly informs how AI-generated contracts will be evaluated.
1. K.L. Katyal v. Dr. Pankaj Kumar (2014) - Intent and Essential Terms
| Aspect | Details |
|---|---|
| Citation | RFA(OS)122/2013 |
| Court | High Court of Delhi |
| Date | 19-03-2014 |
Facts: A contract signed on a blank letterhead was challenged as invalid. The defendant claimed forgery without substantive evidence.
Holding: The Delhi High Court upheld contract enforceability:
"A contract, even if signed on a blank letterhead, is enforceable provided the parties' intent and essential terms are clear. A purported forgery claim must be substantiated with credible evidence; mere allegations are insufficient."
Key Principles:
- Contract validity depends on intent and essential terms, not document format
- The medium of documentation (paper, electronic, AI-generated) is secondary
- Challenges to validity require substantive evidence, not mere allegations
- Courts look to substance over form in contract interpretation
AI Contract Relevance: AI-generated contracts can be valid if they capture parties' intent and contain essential terms. The fact that AI drafted the document does not affect validity if human parties approve and execute it.
2. Classy Mobike Shop v. Yamaha Motor India (2010) - Acceptance Through Conduct
| Aspect | Details |
|---|---|
| Citation | Civil Appeal |
| Court | High Court of Delhi |
| Date | 06-05-2010 |
Facts: A party challenged a dealer agreement's validity, claiming they never received a signed copy.
Holding: The Delhi High Court held the agreement valid:
"Acceptance of goods/services constitutes acceptance of the agreement and its terms, rendering the party bound by it. Such conduct validates an agreement even if a signed copy was not initially returned."
Key Principles:
- Contract can be formed through conduct, not just signature
- Acceptance of benefits implies acceptance of terms
- Parties cannot later deny agreements after receiving benefits
- Electronic/AI-generated terms are binding if accepted through conduct
AI Contract Relevance: When parties use an AI-generated contract and begin performance (paying, delivering, accepting services), they have accepted the contract through conduct - regardless of whether they formally signed the AI output.
3. Ram Sharan Das Batra v. Reliance Webstore (2015) - Letter of Intent vs. Enforceable Contract
| Aspect | Details |
|---|---|
| Citation | CS(OS) 1003/2010 |
| Court | High Court of Delhi |
| Date | 12-05-2015 |
Facts: A plaintiff sought damages based on a document that the defendant claimed was merely a letter of intent, not a binding contract.
Holding: The Delhi High Court clarified the distinction:
"A letter of intent/comfort lacking performance or acceptance cannot be treated as an enforceable contract under Section 8 of the Indian Contract Act. Parties must execute a formal agreement before any obligations arise."
Key Principles:
- Not all documents create binding obligations
- Letters of intent require subsequent formal agreement
- Performance and acceptance are essential for enforceability
- Preliminary documents don't automatically become contracts
AI Contract Relevance: AI may generate drafts that are intended as proposals, not final agreements. Courts will examine whether the document was treated as a binding contract or merely a negotiation tool. Clear labeling of AI outputs as "draft" vs. "final" is essential.
4. Devender Kumar v. Brijesh (2019) - Consideration Requirement
| Aspect | Details |
|---|---|
| Citation | CS(OS) 330/2016 |
| Court | High Court of Delhi |
| Judgment Importance | Land Mark Judgment |
| Date | 16-01-2019 |
Facts: A suit for recovery was filed based on a Memorandum of Understanding. The court examined whether valid consideration existed.
Holding: The Delhi High Court dismissed the suit:
"The MoU was found to be void and unenforceable under Section 25 of the Contract Act. A contract without consideration is void and cannot be enforced."
Key Principles:
- Consideration is mandatory for contract validity (Section 25, Contract Act)
- Even well-drafted documents are void without consideration
- Courts will not enforce agreements lacking mutual exchange
- Form of agreement does not substitute for substance
AI Contract Relevance: AI can draft perfectly formatted contracts, but if they lack consideration (something of value exchanged), they remain unenforceable. Human review must verify that consideration exists, not just that terms are well-drafted.
Summary: Contract Validity Principles for AI-Generated Agreements
| Principle | Judicial Source | AI Contract Application |
|---|---|---|
| Intent and essential terms | Katyal v. Pankaj (2014) | Focus on parties' intent, not AI authorship |
| Acceptance through conduct | Classy Mobike (2010) | Performance validates AI-drafted terms |
| Formal agreement required | Batra v. Reliance (2015) | Label AI outputs as "draft" or "final" |
| Consideration mandatory | Devender v. Brijesh (2019) | Human review must verify consideration exists |
Section 3: Liability Framework
When AI Contracts Go Wrong
Scenario Types:
| Scenario | Issue | Likely Liable Party |
|---|---|---|
| Drafting Error | AI misses critical clause | Lawyer/drafter who approved |
| Hallucinated Terms | AI invents non-existent provisions | User who failed to verify |
| Bias in Terms | AI generates discriminatory clauses | Entity using AI + AI provider |
| Data Breach | Contract data exposed via AI platform | AI provider (under IT Act) |
| Unauthorized Modification | AI alters terms without consent | Depends on access controls |
Proposed Liability Allocation Model
Three-Tier Framework:
Tier 1: User/Deployer Liability
- Primary responsibility for output verification
- Duty to ensure AI used appropriately
- Cannot disclaim liability by blaming AI
Tier 2: AI Provider Liability
- Product liability for defective systems
- Negligence if known risks not disclosed
- Data protection obligations
Tier 3: Shared Liability
- Complex cases where both contribute
- Contribution based on fault assessment
- Indemnity agreements between parties
Existing Legal Provisions
| Law | Applicable Provision | Application to AI Contracts |
|---|---|---|
| Indian Contract Act | Sections 17-19 (Fraud, Misrepresentation) | If AI generates false statements |
| Consumer Protection Act | Product liability | AI as defective product |
| IT Act | Section 43A (Data Protection) | AI platform data breaches |
| Professional Ethics | BCI Rules | Lawyer using AI without verification |
Section 4: Drafting Best Practices
Pre-Drafting Protocol
- Define Scope Clearly: Specify what AI should and shouldn't draft
- Select Appropriate Tool: Use contract-specific AI, not general chatbots
- Input Quality Control: Garbage in = garbage out
- Jurisdiction Specification: Ensure AI knows Indian law context
During Drafting
- Iterative Review: Don't accept first draft
- Clause-by-Clause Verification: Check each provision against intent
- Legal Citation Verification: Confirm any referenced laws/cases exist
- Consistency Check: Ensure no internal contradictions
Post-Drafting Requirements
- Human Review Mandatory: Lawyer must review entire document
- Track Changes: Maintain audit trail of AI vs human contributions
- Disclosure Decision: Consider whether to disclose AI use
- Version Control: Preserve AI output before modifications
Sample Disclosure Clause
DISCLOSURE: Portions of this Agreement were drafted with AI assistance
using [Platform Name]. All AI-generated content has been reviewed,
verified, and approved by [Name], Advocate enrolled with [Bar Council].
The parties acknowledge this disclosure and agree that the reviewing
attorney bears professional responsibility for the document's contents.
Section 5: Sector-Specific Considerations
Employment Contracts
- AI may miss jurisdiction-specific labor laws
- Risk of discriminatory language
- Need for manual review of termination clauses
Real Estate Agreements
- State-specific stamp duty and registration requirements
- RERA compliance for applicable agreements
- Title verification cannot be AI-automated
Technology Contracts
- IP assignment clauses require precision
- Data protection terms need DPDP compliance
- SLA metrics must be verifiable
Financial Contracts
- RBI/SEBI compliance requirements
- FEMA considerations for cross-border
- Interest calculations must be verified
Section 6: Future Regulatory Direction
Expected Developments
- Digital India Act: May include AI-specific provisions
- BCI Guidelines: Expected guidance on AI in legal practice
- SEBI/RBI Circulars: For financial contract automation
- Industry Standards: Self-regulatory frameworks emerging
International Comparison
| Jurisdiction | Approach | India Position |
|---|---|---|
| EU AI Act | Risk-based regulation | Likely to influence |
| US | Sectoral approach | Selective adoption |
| Singapore | Model AI Governance Framework | Potential template |
| UK | Pro-innovation | Balanced approach expected |
Practical Checklist for AI Contract Use
For Legal Professionals
- Verify AI tool is appropriate for contract type
- Ensure client consent for AI-assisted drafting
- Review 100% of AI-generated content
- Verify all legal citations exist
- Check for jurisdiction-appropriate provisions
- Document AI tool used and review process
- Maintain professional liability coverage for AI use
For Businesses
- Establish AI contract policy
- Define approval workflows for AI-drafted contracts
- Train staff on AI limitations
- Maintain human oversight requirements
- Include AI disclosure in contract management records
- Review vendor agreements with AI providers
- Allocate liability in AI service agreements
Conclusion
AI-generated contracts can be valid and enforceable under Indian law when:
- Human parties provide genuine consent
- Contract meets essential validity requirements
- Competent review verifies AI output
- Proper documentation of AI use maintained
The liability framework should allocate responsibility based on:
- Who deployed the AI (user liability)
- Whether the AI was defective (provider liability)
- What verification was performed (professional liability)
As AI contract generation becomes ubiquitous, proactive compliance frameworks will distinguish responsible practitioners from those facing professional consequences.